Eddy Current Evaluation 3-1

Eddy Current Evaluation Page 1 of 38

Error! Reference source not found.

By: Ethan Gros

Idaho National Laboratory Idaho Falls, Idaho 83415

http://www.inl.gov


ACKNOWLEDGMENTS

This research could not have been possible without the assistance of Francine Rice, Katelyn Wachs, Adam Robinson, Steve Marschman, Walter Williams and Michigan State University Professor Dr. Lalita Udpa; as well as the guidance from my mentor James Smith and the support of Idaho National Laboratory and the Department of Energy.


CONTENTS

1. OBJECTIVE OF EDDY CURRENT TESTING...............................................................7-10

1.1 Calibration ...........................................................................................................7-10 1.1.1 Stage One: Zero Setting ............................................................................7-8 1.1.2 Stage Two: Calibration Setting .................................................................9-10

2. FILM THICKNESSES MEASUREMENT: ACCURACY, REPEATABILITY .................. 12-15

2.1 Objective ................................................................................................................ 12

2.2 Description of Sample............................................................................................. 12

2.3 Description of Test Procedure ................................................................................ 12

2.4 Results & Disscusion ......................................................................................... 12-14 2.4.1 Day A Test ............................................................................................. 12-13 2.4.2 Day B Test .................................................................................................. 14

2.5 Conclusion.............................................................................................................. 15

3. EFFECT OF PROXIMITYTO EDGE ........................................................................... 17-18

3.1 Objective ............................................................................................................... 17


3.3 Description of Test Procedure ............................................................................... 17

3.4 Results & Disscusion ......................................................................................... 17-18

3.5 Conclusion.............................................................................................................. 18

4. EFFECT OF CONDUCTIVITY OF MATERIALS ...........................................................20-2

4.1 Objective ................................................................................................................ 20


4.3 CASE I .............................................................................................................. 20-21 4.3.1 Description of Test Procedure ................................................................20-21 4.3.2 Results & Disscusion .................................................................................. 21 4.3.3 Conclusion .................................................................................................. 21

4.4 CASE II ............................................................................................................. 21-26 4.4.1 Description of Test Procedure ................................................................21-22 4.4.2 Description of Sample ................................................................................. 22 4.4.3 Conductivity Test 1 – Materials with similar conductivity ........................22-23

4.4.3.1 Conclusion .......................................................................................... 23 4.4.4 Conductivity Test 2 - Day B ....................................................................23-24

4.4.4.1 Conclusion .......................................................................................... 24 4.4.5 Conductivity Test 3 – Materials with dissimilar conductivity ......................... 25

4.4.5.1 Description of Test Procedure ............................................................. 25 4.4.5.2 Conclusion .......................................................................................... 25

4.4.6 Conductivity Test 4 – Calibration on Stainless Steel ................................... 26

4.4.6.1 Description of Test Procedure ............................................................. 26 4.4.6.2 Conclusion .......................................................................................... 26


5. EFFECT OF SURFACE FINISH ................................................................................. 28-32

5.1 Objective ................................................................................................................ 28


5.3 Description of Test Procedure ........................................................................... 28-29


5.5 Conclusion.............................................................................................................. 31

5.6 Additional Surface finish Test ............................................................................ 31-32 5.6.1 Description of Test Procedure ..................................................................... 31 5.6.2 Results & Disscusion .................................................................................. 32 5.6.3 Conclusion .................................................................................................. 32

6. EFFECT OF CABLE ................................................................................................... 34-36

6.1 Objective ............................................................................................................... 34


6.3 Description of Test Procedure ........................................................................... 34-35


6.5 Conclusion.............................................................................................................. 36

7. RECOMMENDATIONS .................................................................................................... 38

FIGURES

Figure 1. Front panel of Fischer FMP 40 handle gauge with “Zero” key marked in red... .............7

Figure 2. Experimental set up showing probe and Al T6 test plate with (a) probe raised (b) pressed on top of sample. . .....................................................................................8

Figure 3. Fischer FMP 40 handle gauge (a) with “10” mark in blue indicates the number

of times the probe was pressed on the Al sample and the “Enter” key marked in green. (b) Screen with “Normalization finished successfully” message. . ......................8

Figure 4. Fischer FMP 40 gauge (a) with the calibration screen up after pressing the “CAL” button marked in red. (b) the “10” mark in blue which indicates the

number of times the probe was pressed on the Al sample and “Enter” key marked in green............................................................................................................9

Figure 5. Fischer FMP 40 gauge (a) with the calibration film 1 size (5.46 µm) marked in teal (b) with “10” mark in blue indicates the number of times the probe was pressed on the Al sample and “Enter” key marked in green... .......................................9

Figure 6. Fischer FMP 40 gauge (a) Fischer FMP 40 gauge (a) with the calibration film 1 size (49.96 µm) marked in teal. (b) the “10” mark in blue which indicates the

number of times the probe was pressed on the Al sample and “Enter” key marked in green. The message (c) “Corrective Calibration finished

successfully” appears on the screen indicating the calibration was successfully. . ..................................................................................................................................10

Figure 7. Fischer Test Films.. ....................................................................................................12

Figure 8. This image is the tip of the eddy current probe next to a ruler. . .................................18


Figure 9. These are the two test plates with different surface finishes that were used... ............29

Figure 10. This image conveys cable to the eddy current probe coiled......................................34

Figure 11. This image depicts the cable to the eddy current probe kinked. ...............................35

TABLES

Table 1. Mean values of measured thickness along with percent error on Day A..... .................13

Table 2. Mean values of measured thickness along with percent error on Day B..... .................14

Table 3. Mean and standard deviation of thickness measured at each (X, Y) position of probe.…......................................................................................................................17

Table 4. Mean and standard deviation of thickness measured at each (X, Y) position of probe.…......................................................................................................................18

Table 5. Conductivity values of the alloys used...... ...................................................................20

Table 6. Mean values of film thickness measured on different alloys. CASE I …. .....................21

Table 7 Conductivity values of the alloys the probe was calibrated on. .…. ...............................22

Table 8 Mean values of thickness measured on each material with different conductivity value along with percent error. The probe was calibrated on the Al 6061

T6..…. ........................................................................................................................23

Table 9 Mean values of thickness measured on each material with different conductivity value along with percent error. The probe was calibrated on the Al 6061 T6…. ..........24

Table 10 Mean values of the measured film on materials with different conductivity values. The probe was calibrated on Al 6061-T6 plate..…. .........................................25

Table 11 Mean values of the measured film thickness on base materials of different conductivity values. The probe was calibrated on Stainless Steel plate.…..................26

Table 12 Measured values of film thickness on different surface finish along with standard deviation, and the percent error of measurements. (Actual Value = 77.10 µm)…................................................................................................................30

Table 13 Measured values of film thickness on different surface finish along with standard deviation, and the percent error of measurements …. ..................................31

Table 14 Effect of surface finish on film thickness measurement (Actual film thickness in all cases = 12.22 µm) …. ............................................................................................32

Table 15 Case I, II and III results of thickness measurements along with % error.…. ................36

GRAPH

Graph 1. Conductivity Percent (%) vs Percent Error (%)...... .....................................................24


[This page is meant to be blank]


1. OBJECTIVE OF EDDY CURRENT TESTING

Eddy-current techniques are widely used in industry to measure the thickness of non-conductive films on a metal substrate. This is done using a system whereby a coil carrying a high-frequency alternating current is used to create an alternating magnetic field at the surface of the instrument's probe. When the probe is brought near a conductive surface, the alternating magnetic field will induce eddy currents in the conductor. The substrate characteristics and the distance of the probe from the substrate (the coating thickness) affect the magnitude of the eddy currents. The induced currents load the probe coil affecting the terminal impedance of the coil. The measured probe impedance is related to the lift off between coil and conductor as well as conductivity of the test sample. For a known conductivity sample, the probe impedance can be converted into an equivalent film thickness value.

The eddy-current test is performed using a commercially available, hand-held eddy-current probe (ETA3.3H spring-loaded eddy probe running at 8 MHz) that comes with a stand to hold the probe. The stand holds the probe and adjusts the probe on the z-axis to help position the probe in the correct area as well as make precise measurements. The signal from the probe is sent to a hand-held readout, where the results are recorded directly in terms of liftoff or film thickness.

Understanding the effect of certain factors on the measurements of film thickness, will help to evaluate how accurate the ETA3.3H spring-loaded eddy probe is at measuring film thickness.

This report documents the results of a study conducted to evaluate the effect of a number of factors such as i) calibration, ii) conductivity, iii) edge effect iv) surface finish of base material and v) cable condition.

1.1 CALIBRATION 1.1.1 Stage One: Zero Setting

First the meter readout was set by pressing the button ZERO this sets the thickness of film on base material as zero(as indicated in Figure 1).

This is the initial screen after pressing the “zero” button.

Figure 1. Front panel of Fischer FMP 40 handle gauge with “Zero” key marked in red.


The probe was placed in a stand, so the probe was perpendicular to the surface of the metal

test plate. Figure 2(a) shows the initial set up with the Aluminum T6 – Test Plate on the stand.

The lever was pulled in order to lower the probe onto the Aluminum T6 – Test Plate pressing the

probe against the Aluminum Test Plate as indicated in Figure 2(b).

Aluminum T6 – Test Plate

Lever to lower probe

Pressed probe

The probe was pressed against the plate 10 times (or as many times as the experiments required) and then ENTER was pressed as indicated in Figure 3(a). When the information is

entered, the message “Normalization finished successfully” appears on the screen indicating the zero calibration was successfully as indicated in Figure 3(b).

(a) (b)

(a) (b)

Figure 2. Experimental set up showing probe and Al T6 test plate with (a) probe raised (b)

pressed on top of sample.

Figure 3. Fischer FMP 40 handle gauge (a) with “10” mark in blue indicates the

number of times the probe was pressed on the Al sample and the “Enter” key

marked in green. (b) Screen with “Normalization finished successfully” message.


1.1.2 Stage Two: Calibration Setting

First the probe was calibrated by pressing the button CAL which starts the calibration process as indicated in Figure 4(a).The probe was then pressed against the plate 10 times (or as many times as the experiments required) and ENTER was pressed as indicated in Figure 4(b). A

calibration film 1 of thickness 5.46 µm was then placed on the aluminum plate. Then the output of the gauge was adjusted according to the known film thickness. Figure 5(a) indicates the calibration film 1 thickness, which in this case was set to 5.46 µm.

Again probe was pressed against the plate 10 times (or as many times as the experiment requires) and then ENTER was pressed (as indicated in Figure 5(c)). Figure 6(a) indicates the

calibration film 2 thickness which in this case was 49.96 µm. This film thickness can be varied to any value. Again probe was pressed against the plate 10 times or as many times as the experiments require and then ENTER was pressed (as indicated in Figure 6(b)). When the last

10 data points were entered, the message “Corrective Calibration finished successfully” appears on the screen indicating the calibration was successful (as indicated in Figure 6(c)).

Figure 4. Fischer FMP 40 gauge (a) with the calibration

screen up after pressing the “CAL” button marked in red.

(b) the “10” mark in blue which indicates the number of

times the probe was pressed on the Al sample and

“Enter” key marked in green.

Figure 5. Fischer FMP 40 gauge (a) with the

calibration film 1 size (5.46 µm) marked in

teal (b) with “10” mark in blue indicates the

number of times the probe was pressed on

the Al sample and “Enter” key marked in

green.

Figure 4(a) Figure 4(b) Figure 5(a)

This is the initial screen after pressing the CAL

button.


Figure 5(b) Figure 6(a) Figure 6(b) Figure 6(c)

Figure 6. Fischer FMP 40 gauge (a) Fischer FMP 40 gauge (a) with the calibration film 1 size

(49.96 µm) marked in teal. (b) the “10” mark in blue which indicates the number of times the probe

thickness was measured. (c) “Corrective Calibration finished successfully” appears on the screen




2. FILM THICKNESSES MEASUREMENT: ACCURACY, REPEATABILITY

2.1 OBJECTIVE

This section documents the results of thickness measurements on different film thicknesses. The objective of this experiment was to determine the performance of the instrument in predicting the thickness of films of known thickness values with respect to accuracy and repeatability of measurements.

2.2 DESCRIPTION OF SAMPLES.

Different aluminum alloy samples were used as base material in this test to determine the repeatability and accuracy of the film thickness measurement. The films used in this study were provided by the Fischer Company as standards and are shown in Figure 7.

Figure 7. Fischer Test Films

The probe was calibrated with a plate of known conductivity and film thicknesses were measured on the same plate.

2.3 DESCRIPTION OF TEST PROCEDURE

The probe was placed in the stand perpendicular to the surface of a specific Al alloy as the

metal test plate. The probe was then calibrated as described earlier.

Then the probe was used to measure several different film thicknesses on top of the specific

alloy on which the Eddy Current Fischer Probe was calibrated.

2.4 RESULTS & DISCUSSION

2.4.1 Day A Test

The probe measured each film thickness on each alloy three times to get a sample population of

3. Then the three measurements were averaged to get a mean value for each film thickness (as

depicted in Table 1). The percent error was calculated using this formula |𝐴𝑐𝑡𝑢𝑎𝑙 − 𝑀𝑒𝑎𝑢𝑠𝑢𝑟𝑒𝑑|

𝐴𝑐𝑡𝑢𝑎𝑙× 100% to

compare the percent error with respect to different film thickness values.


Table 1. Mean values of measured thickness along with percent error on Day A.

Key

Percent error 5% or higher



2.4.2 Day B Test

The probe measured each film on each alloy three times to get a sample population of 3. Then

the three measurements were averaged to get a mean value and percent error for each film

thickness (as depicted in Table 2).

Key




Table 2. Mean values of measured thickness along with percent error on Day B.


2.5 CONCLUSION

The data collected on Day A has 6 thickness values that have > 5% error and of these 6 data

points 5 of them are with film thickness < 12.22 µm. Also of these 5 data points 3 of them have

> 15% error. Experiments performed on Day B included additional base material alloys. Data

collected on Day B has 29 thickness values that have > 5% error. Of these 29 points 21 of them

are with film thickness < 12.22 µm. Also there are 14 values that have > 15% error and of these

14 data points 9 are with film of thickness < 5.52 µm and 12 of them are with film of thickness <

12.22 µm.

Therefore it can be concluded that the smaller the film thickness value the larger the percent or

relative error. However, the data is repeatable as long as the probe is calibrated on the same

sample on which the film thickness was measured.




3. EFFECT OF PROXIMITY TO EDGE

3.1 OBJECTIVE

This section documents the results of edge effect on eddy current testing. The objective of this test was to determine whether the proximity of an edge affects the measurements of film thickness.

3.2 DESCRIPTION OF SAMPLES

This test was performed on Aluminum 6061 T6 – Test Plate. The Eddy Current Fischer Probe

was first calibrated on the same base plate. The test film thickness used was 77.10 µm


In order to determine the edge effects the probe was first calibrated and placing a film of known

thickness, the experiment consisted of measuring the thickness as the probe was moved from

the middle of the plate out to the edge of the plate.

The probe was moved in the X direction in steps of 5 mm from the center of the test plate

taking 10 data points at each new coordinate (ex. (45,12) (40,12)…) until the probe was

positioned 5 mm from the edge of the test plate (5mm,12mm). The Y-coordinate was kept

constant at 12 mm.

Next, the Y coordinate of probe was varied in steps of 2 mm from the center of the test

plate (ex. (50,10) (50,8)…) until the probe reached 2 mm from the edge of the test plate

(50mm,2mm). The X-coordinate was kept constant at 50 mm.

At each (X,Y) position of the probe, average of 10 measurements values and standard deviation

were calculated. These results are summarized in Table 3 for X direction movement and Table

4 for Y direction movement of the probe.

3.3 RESULTS & DISSCUSSION

Table 3. Mean and standard deviation of thickness measured at each (X, Y) position of probe.


3.4 CONCLUSION

The tip of the probe is about 3mm in diameter as indicated in Figure 8 and therefore any data

collected 2mm or 4mm from the edge is likely to be in error as seen in the values measured at

the coordinates (50,4) and (50,2). From the results presented in Tables 3 and 4, it is concluded

that the ETA3.3H spring-loaded eddy current probe is not significantly affected by proximity to

edges as long as the probe is at least 4mm from the edge.

Figure 8. This image is the tip of the eddy current probe next to a ruler.

Table 4. Mean and standard deviation of thickness measured at each (X, Y)

position of probe.




4. EFFECT OF CONDUCTIVITY OF MATERIALS

4.1 OBJECTIVE

This section documents the results of thickness measurements using eddy current testing on materials of conductivity different from that of the calibration sample conductivity. The objective of this test was to determine whether conductivity of the material affects the measurements of film thickness.

Two cases were considered as below:

*CASE I: CALIBRATION SAMPLE IS SAME AS SAMPLE USED FOR THICKNESS

MEASUREMENT

*CASE II: CALIBRATION SAMPLE IS DIFFERENT FROM THE SAMPLE USED FOR

THICKNESS MEASUREMENT

4.2 DESCRIPTION OF SAMPLES

In this test we used samples of different conductivity as depicted in Table 5.

Table 5. Conductivity values of the alloys used.

Three different film thicknesses, namely, 11.42 µm, 22.00 µm and 49.54 µm were used to test the effect of conductivity on thickness measurement.

4.3 CASE I: CALIBRATION SAMPLE IS SAME AS SAMPLE USED FOR THICKNESS

MEASUREMENT

4.3.1 DESCRIPTION OF TEST PROCEDURE

The probe was placed in the stand perpendicular to the surface of the metal test plate. The

probe was then calibrated on a specific sample.


The probe was calibrated on each alloy before taking the film thickness measurements. As described earlier, 10 measurements were taken for each film thickness. These 10 measurements were then averaged to get mean value as shown in Table 6.

4.3.2 RESULTS & DISSCUSION

4.3.3 CASE I: CONCLUSION

The probe produces reasonably accurate measurement of film thicknesses on top of any base

material as long as the probe was calibrated on the same material. The data obtained with

Copper, Tungsten and Stainless steel had higher error in measurement (the data highlighted in

red), however the data variation was about ±0.60 µm and in one extreme case ±0.92 µm, which

translates to 2% or less error.

4.4 CASE II: CALIBRATION SAMPLE IS DIFFERENT FROM THE SAMPLE USED FOR

THICKNESS MEASUREMENT


In this study, the thickness was measured by placing the film on samples of conductivity

different from that of the calibration sample. The Eddy Current Fischer Probe on was calibrated

on a specific material (Aluminum 6061 T6 – Test Plate).

Table 6. Mean values of film thickness measured on different alloys. CASE I


Then a specific test film was placed on top of different base material plates and its thickness

was measured to determine whether the conductivity of base material affects the film thickness

measurements.

4.4.2 DESCRIPTION OF SAMPLES

In this test we used samples of different conductivity values as depicted in Table 7.

Table 7. Conductivity values of the alloys the probe was calibrated on.

Five different film thicknesses, namely, 5.52 µm, 11.42 µm, 22.00 µm, 49.54 µm, and 77.10 µm, were used.

4.4.3 CONDUCTIVITY TEST 1 – Materials with similar conductivity

The probe was calibrated on Aluminum 6061 T6 – Test Plate (row highlighted in green) and

then 10 measurements were taken for the film thickness of 77.10 µm. These 10 measurements

then were averaged to get a mean value of thickness measured on each material with a different

conductivity. The percent error column was calculated using this formula |𝐴𝑐𝑡𝑢𝑎𝑙 − 𝑀𝑒𝑎𝑢𝑠𝑢𝑟𝑒𝑑|

𝐴𝑐𝑡𝑢𝑎𝑙× 100% .


4.4.3.1 CONCLUSION

Although the conductivity of calibration sample was different from that of the base material used when measuring the film thickness, there seems to be a large (>4%) error in only two samples, namely, aluminum 6061 T6511 and Al 6061 Hipped. The conductivity of these two materials were not available and it is possible that the error is due to difference in conductivity of the calibration sample and base material. When there is not a significant change in conductivity values among the materials in table 8 there is not a significant error in thickness values when the probe was calibrated on Al 6061 T6 and thickness is measured on the other alloys .

4.4.4 CONDUCTIVITY TEST 2 – DAY B

The probe was calibrated on Aluminum 6061 T6 – Test Plate with a straight cable (row

highlighted in green) and then 10 measurements were taken for the film thickness of 12.22 µm

and a straight cable. These 10 measurements then were averaged to get a mean values as

conveyed in table 9. The probe was calibrated only on Aluminum 6061 T6 – Test Plate. Then

the 12.22 µm thick film was measured on each of the other alloys.

Table 8. Mean values of thickness measured on base materials with different conductivity value along

with percent error. The probe was calibrated on the Al 6061 T6.


4.4.4.1 CONCLUSION

The results in Table 9 are at odds with what was observed earlier. The errors are relative larger

in all these measurements. Some of the base material conductivity values that are largely

different from that of calibration sample are seen to have larger error. However there is no

observable correlation between conductivity difference and error in predicted film thickness.

On the first row, the alloy Al6016-T0 has the largest error namely, 18%, whereas the

conductivity difference between this sample and the calibration sample is roughly 10%. The

randomness of observations in Table 9 is rather troubling and needs to be investigated further.

Table 21

Table 9. Mean values of thickness measured on each material with different conductivity value

along with percent error. The probe was calibrated on the Al 6061 T6

Graph 1. Conductivity Percent (%) vs Percent Error (%).


4.4.5 CONDUCTIVITY TEST 3 – Materials with dissimilar conductivity

4.4.5.1 DESCRIPTION OF TEST PROCEDURE

In this test, the Eddy Current Fischer Probe was calibrated on an Aluminum 6061 T6 – Test

Plate (highlighted). The film thickness was the measured on completely different metals such as

stainless, steel, copper and tungsten. The test film thicknesses used were 11.42 µm, 22.00

µm and 49.54 µm. There were 10 measurements taken for each film thickness. These 10

measurements were then averaged to get a mean value as listed in table 10.

4.4.5.2 CONCLUSION

From Table 10, we observe that the measurements of film thickness on top of copper and the

stainless steel, were in significant error. The stainless steel has lower conductivity value

than that of calibration sample and the measured thickness values were seen to be consistently

larger than the true film thickness. On the other hand copper has a higher conductivity

value than aluminum and the measured thickness values are consistently smaller than the

true film thickness. This is to be expected since higher conductivity value of base material

results in higher eddy current probe signal that can be translated as lower liftoff and hence

smaller film thickness and vice versa.

Table 10. Mean values of the measured film on materials with different conductivity values. The probe

was calibrated on Al 6061-T6 plate.


4.4.6 CONDUCTIVITY TEST 4 –Calibration on Stainless Steel

4.4.6.1 DESCRIPTION OF TEST PROCEDURE

In this test, the Eddy Current Fischer Probe was calibrated on a Stainless Steel with

calibrations films of thickness 5.46 µm and 11.42 µm. Then a test film of thickness 5.52 µm

was placed on Al6061-T6 and Copper plates. In each case 10 measurements taken and

averaged to get a mean value as listed in table 11.

4.4.6.2 CONCLUSION

The measurements of film thickness on copper and Al 6061-T6, which both have conductivity

value larger than stainless resulted in negative film thickness output by the instrument, when it

was first calibrated on stainless steel. Therefore it could be hypothesized that if the calibration

sample conductivity is lower than the base material conductivity, the instrument outputs

negative thickness values!! This is clearly a problem with the current instrument.

Table 11. Mean values of the measured film thickness on base materials of different conductivity

values. The probe was calibrated on Stainless Steel plate.




5. EFFECT OF SURFACE FINISH

5.1 OBJECTIVE

This section documents the results of conducting the thickness measurement test on samples with different surface finishes. The objective of this test was to determine whether the surface finish affects the measurements of film thickness.

5.2 DESCRIPTION OF SAMPLE

The experiment used two metal test plates. One test plate had six panels with different machined surface finishes and the second test plate had nine different machined surface finishes as shown in Figure 9(a) and 9(b).

The G-6 Grit-Blast Microfinish Comparator Surface Roughness Scale (Surface

Roughness Scale) (Figure 9(a)) provides industry with established flat surface roughness

specimens for visual and actual comparison. It is used in specifying and controlling surface

roughness when a product having the appearance of typically machined surfaces is required.

This scale uses a reproduction of accurate machined surfaces measured in microinches. The

numbers engraved alongside each surface are the average deviation from the mean surface

expressed in microinches.

The E-9 Microfinish Comparator Surface Roughness Scale (Microfinish Comparator

Scale) (Figure 9(b)) is made by a dual electroforming process wherein nickel is

electrodeposited to provide an exact reproduction in intricate detail. The surfaces used in this

scale are reproductions of accurately electrical discharge machined surfaces on oil-hardening

tool steel measured in microinches. The numbers engraved alongside each surface are the

average deviation from the mean surface expressed in microinches.



probe was then calibrated on top of the metal test plate on the smoothest surface. This

calibration was done using 22.10 µm and 256.80 µm thick films.

Then a test film of thickness 77.10 µm was placed on top of the lowest surface finish value. The

probe was lowered onto the surface of the plate, so that the probe was completely pressed on

the plate and film thickness was measured. This measurement was repeated 9 times on the

same spot with the same film thickness, so that a total of 10 values were averaged and

recorded. This measurement was repeated with the same film on panels of different surface

finish on the G-6 and E-9 samples.


The blue circle represents

the film with the known

thickness 77.10 µm.

The red circle represents

the relative location that

the eddy current probe

touched the surface.

Figure 9 (a & b). These are the

two test plates with different

surface finishes that were used.

Figure 9(a) Figure 9(b)



Table 12 presents the mean value of the thickness value obtained with different surface finish in

the two metal test plates. The true film thickness is 77.10 (µm). Also the percent error was

calculated using the formula |Actual − Meausured|

Actual× 100% , where “Actual” value is 77.10 µm and

“Measured” is the output of the instrument. The percent error in film thickness measurement for

each surface finish is provided in the last column

Table 13 presents the corresponding film thickness values measured (average of 10

measurements, standard deviation and percent error) on the second metal plate.

Table 12. Measured values of film thickness on different surface finish along with standard

deviation, and the percent error of measurements. (Actual Value = 77.10 µm)


5.5 CONCLUSION

The surface finish on which the thickness measurement is made has significant effect on the output of ETA3.3H eddy probe system. Table 12 indicates that as the surface finish becomes coarser, the error in measured film thickness value gets larger. Table 13 also indicates that the error is small for good surface finish and increases with coarser grain finish. Hence, for the instrument to provide an accurate estimate of the oxide layer thickness, a good surface finish should be ensured. This conclusion is to be expected since the measurements are made at very high frequency (8MHz) and eddy current skin depth being very small at this frequency carries more information on the surface condition. Clearly, the instrument is very sensitive to surface condition of the sample on which measurement is made.

5.6 ADDITIONAL SURFACE FINISH TEST


In this test, the calibration and film thickness measurement is made on the same panel, i.e. with the same surface finish.

The Eddy Current Fischer Probe was calibrated on the G-6 Grit-Blast Microfinish Comparator

Surface Roughness Scale (Surface Roughness Scale) at the 32 µin surface finish. A test film

of 12.22 µm thickness was placed on top of the 32 µin surface finish and tested 10 times.

Table 13. Measured values of film thickness on different surface finish along with standard

deviation, and the percent error of measurements


Next, the Eddy Current Fischer Probe was re-calibrated on the G-6 Grit-Blast Microfinish

Comparator Surface Roughness Scale at the 1000 µin surface finish. A test film of 12.22 µm

thickness was placed on top of the 1000 µin surface finish value and tested 10 times.

Next, the same two tests were conducted on the E-9 Microfinish Comparator Surface

Roughness Scale (Microfinish Comparator Scale).

The Eddy Current Fischer Probe was calibrated on the E-9 Microfinish Comparator Surface

Roughness Scale at the 16 µin surface finish. A test film of 12.22 µm thickness was placed on

top of the 16 µin surface finish and tested 10 times.

Next, the Eddy Current Fischer Probe was re-calibrated on the E-9 Microfinish Comparator

Surface Roughness Scale at the 250 µin surface finish. A test film of 12.22 µm thickness was

placed on top of the 250 µin surface finish value and tested 10 times.

5.6.2 RESULTS & DISSCUSION

Table 14 presents the 10 measurements values averaged to get a mean value for each surface

finish/ film thickness combination. Also the percent error was calculated using this formula |Actual − Meausured|

Actual× 100% to compare the percent error of due to surface finish.

5.6.3 CONCLUSION

The data collected on G-6 Grit-Blast Microfinish Comparator Surface Roughness Scale

supports the idea “that as surface roughness increases, so does the error in film thickness

measurements”.

Table 14. Effect of surface finish on film thickness measurement (Actual film thickness in all cases =

12.22 µm)




6. EFFECT OF CABLE CONDITION

6.1 OBJECTIVE

This section documents the results of measurements made with a coiled cable. The objective of this study was to determine whether capacitive effects of coiled cable will affect the performance of the eddy current probe.

6.2 DESCRIPTION OF SAMPLE

Three different cable configurations were conducted in order to determine whether the cable effected the film thickness measurements.


CASE I: COILED CABLE


probe was then calibrated on top of a specific alloy.

In order to perform the test the Eddy Current Fischer Probe was calibrated with a straight cable

on a specific material (Aluminum 6061 T6 – Test Plate).

Then the cable was coiled (as depicted in Figure 10) and a specific test film was placed on

base materials with different conductivity values. The test film thickness was then measured

with the eddy current probe.

CASE II: KINKED CABLE

The Eddy Current Fischer Probe was calibrated with a straight cable on Aluminum 6061 T6 –

Test Plate.

Coiled Cable

Figure 10. Eddy current probe measurement with a coiled cable


A test film was placed onto materials with different conductivity values. The test film

thickness was then measured with the eddy current probe connected to the “kinked” cable

(as depicted in Figure 10) to determine whether the test film thickness measurement was

affected by the cable configuration.

CASE III: STRAIGHT CABLE

The probe was calibrated on Aluminum 6061 T6 – Test Plate with a straight cable (row

highlighted in green) and then 10 measurements were taken for the film thickness of 12.22 µm

and a straight cable. These 10 measurements then were averaged to get a mean values as

conveyed in table 15. The probe was calibrated only on Aluminum 6061 T6 – Test Plate.

Then the 12.22 µm film was measured on each of the other alloys.


The probe was calibrated on Aluminum 6061 T6 – Test Plate (row highlighted in green). The

test film of true thickness 11.42 µm or 12.22 µm was used on different base materials. 10

measurements were averaged to get mean values as listed in table 15. The percent error (a)

column was calculated using the formula |𝐴𝑐𝑡𝑢𝑎𝑙 − 𝑀𝑒𝑎𝑢𝑠𝑢𝑟𝑒𝑑|

𝐴𝑐𝑡𝑢𝑎𝑙× 100% .

Kinked Cable

Figure 10. Eddy current probe measurement with a kinked cable


Table 15. Case I, II and III results of thickness measurements along with % error

6.5 CONCLUSION

The results of this study are not very conclusive. There is no clear correlation between the coil

condition and the error in thickness value measured by the instrument. It was expected that the

straight cable would present lowest error in measurement. However, there is no measurable

correlation between the error and cable condition. Further there is also no correlation between

the error in thickness value and base material conductivity.

When the base material is Al6061-T0 tempered alloy, the error in thickness is consistently high.

Although in Table 8, the measurement on Al 6061-T0 produced a small % error. The reason for

this observation needs to be further investigated. At this point, this could be due to repeatability

of performance of the instrument.




7. RECOMMONDATIONS

In summary, this report documents a study of different parametric effects on film thickness values measured by the ETA3.3H eddy current probe. The parameters considered are base material conductivity, edge effect, surface finish and cable condition. A major problem that was observed is in repeatability of the measurements. 1. The performance of the instrument with respect to edge effect is consistent with physics of eddy currents. At 8 MHz, the fields are fairly localized and footprint of probe is very small. Unless the probe is on top of an edge or < 2 mm from edge, the results are not affected by edge. 2. The effect of surface condition of the material is consistent with physics of eddy currents. At very high frequency, the probe is very sensitive to surface condition and hence thickness measurements. 3. The capacitive effects at 8 MHz seems to be much smaller and did not produce significant error. 4. Conductivity of the calibration sample relative to that of base material is expected to affect thickness measurement. This was clearly observed when the materials chosen for calibration is Aluminum alloy and base materials were copper and stainless steel. But when differently treated alloys of Aluminum were chosen as base material, no consistent correlation was observable. It is possible that the conductivity values used from handbook maybe different from true conductivity of the material. 5. Another factor that was not investigated is the effect of temperature. It is well-known that conductivity of most materials is a function of temperature and hence if we have an accurate estimate of this dependence and an accurate knowledge of the temperature of measurement , the effect of temperature can be compensated.

Documents

Eddy Current Evaluation 3-1