1. William Corbett, Carly McGee and Jason Coley KTA-Tator, Inc.
Too Deep or Too Shallow… Can Surface Profiles Be Changed by Additional Abrasive Blast Cleaning?
Author & Presenter: William Corbett, PCS
Co-Author: Carly McGee, PCS
William Corbett, Carly McGee and Jason Coley
KTA-Tator, Inc.
© 2007 Microsoft Corporation. All rights reserved. Microsoft, Windows, Windows Vista and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries.
The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

2. Origin & Learning Outcomes
Research Conducted in 2016
Technical Paper Presented at SSPC 2017 Conference
Abbreviated Version of Technical Paper Published by JPCL (February 2019)
Learning Outcomes:
Describe the process of generating a surface profile;
Recognize the impact of insufficient or excessive surface profile on coating performance;
Describe the process of increasing or decreasing surface profile; and
Recognize the impact of reducing surface profile on peak density/count

3. Background & Problem Statement
Coating specifications invoking abrasive blast cleaning generally contain surface profile requirements
Imparting surface profile enhances mechanical adhesion and helps resist undercutting corrosion
Insufficient or excessive surface profile can be problematic
Measurement of surface profile is a frequent hold point inspection

4. Background & Problem Statement
Several factors influence surface profile depth
Nozzle distance and angle
Compressed air volume & pressure
Abrasive type, size & hardness
Substrate type & hardness
Presence of pre-existing profile
Preparation of a test area can be advantageous

5. Background & Problem Statement
Some evidence that a shallow surface profile can be deepened using a larger abrasive
Re-profiling presents challenges for blaster and QC
Re-profiling doesn’t change appearance
Confirming 100% of the surface has been re-prepared is difficult

6. Background & Problem Statement
Excessive profile caused by:
Existing surface profile from previous work
“Softer” steel
Selection of an abrasive that is too large
Can surface profile depth be reduced by re-blast cleaning?
Does parallel re-blast work?
Is peak density/count reduced?
Re-profiling presents challenges for blaster and QC

7. Experimental Design – Phase 1
Included two abrasive types
Mineral (Garnet)
Ferrous Angular (Steel grit)
Effect of Re-Blast Cleaning Existing Shallow and Deep Surface Profile
Plate No./
Abrasive
Quad 1 2 3 4
Re-blast
(Larger1)
(Larger2)
(Smaller1)
(Smaller2)
1, 2, 3 (Mineral)
1-mil MS
Quad 2-4
4, 5, 6 (Ferrous)
7, 8, 9 (Mineral)
5-mil
10, 11, 12 (Ferrous)
MS: Mill Scale
1 Used a mineral abrasive to generate a nominal 2-3 mil surface profile
2 Used a ferrous, angular abrasive to generate a nominal 2-3 mil surface profile

8. Experimental Design – Phase 2
Included two abrasive types
Mineral (Garnet)
Ferrous Angular (Steel grit)
Effect of Parallel Re-Blast Cleaning of an Existing Deep Surface Profile
Plate No./
Abrasive
Quad 1 2 3 4
Re-blast1
(Parallel; ~30°)
Re-blast2
13, 14, 15
(Mineral)
5-mil
Quad 1-2 only
16, 17, 18
(Ferrous)
1 Used a mineral abrasive
2 Used a ferrous, angular abrasive

9. Experimental Design – Phases 1&2
Methods Used to Characterize the Surfaces
ASTM D4417
Method B (depth micrometer)
Method C (replica tape/micrometer)
Peak Density (optical grade replica tape with reader)
ASTM D7127
Stylus Instrument for Peak Count (Rpc)
Three-dimensional microscopy

11. Experimental Design – Phases 1&2
Frequency of Measurements
ASTM D4417
Method B (depth micrometer): 10 per area, averaged
Method C (replica tape/micrometer): 2 per area, averaged
Peak Density (optical grade replica tape with reader)
2 per area, averaged
ASTM D7127
Stylus instrument for Peak Count (Rpc)
Five traces per area
Three-dimensional microscopy
One image per area

12. Test Results Variable No. 1
Shallow profile (e.g., 1-mil) with Garnet. Can it be increased with a larger abrasive?
Technique
Attributes of Initial Blast on Mill Scale (Q1)
Attributes of Second Blast on Mill Scale (Q4)
Attributes of Re-Blast
(Q2 & Q3)
Depth Micrometer
1.5 mils
2.2 mils
Replica Tape
2.0 mils
3.0 mils
2.9 mils
Peak Density
17/mm2
14/mm2
16/mm2
Peak Count
83/cm
67/cm
69/cm
Characteristic
Yes No
% Change
Did re-blast increase surface profile? X
+32%1 & +31%2
Did re-blast change the peak density?
-6%
Did re-blast change the peak count?
-17%
1Depth Micrometer 2Replica Tape

13. Test Results Variable No. 1
Initial roughness with smaller abrasive (Q1)
Roughness generated by re-blast with larger abrasive (Q2)

14. Test Results Variable No. 2
Shallow profile (e.g., 1-mil) with Steel Grit. Can it be increased with a larger abrasive?
Technique
Attributes of Initial Blast on Mill Scale (Q1)
Attributes of Second Blast on Mill Scale (Q4)
Attributes of Re-Blast
(Q2 & Q3)
Depth Micrometer
1.8 mils
2.6 mils
2.5 mils
Replica Tape
2.7 mils
Peak Density
20/mm2
13/mm2
12/mm2
Peak Count
116/cm
76/cm
77/cm
Characteristic
Yes No
% Change
Did re-blast increase surface profile? X
+28%1 & +33%2
Did re-blast change the peak density?
-40%
Did re-blast change the peak count?
-34%
1Depth Micrometer 2Replica Tape

15. Test Results Variable No. 2
Initial roughness with smaller abrasive (Q1)
Roughness generated by re-blast with larger abrasive (Q2)

16. Test Results Variable No. 3
Deep profile (e.g., 5-mil) with Garnet. Can it be reduced with a smaller abrasive (nozzle perpendicular)?
Technique
Attributes of Initial Blast on Mill Scale (Q1)
Attributes of Second Blast on Mill Scale (Q4)
Attributes of Re-Blast
(Q2 & Q3)
Depth Micrometer
4.6 mils
2.1 mils
2.8 mils
Replica Tape
4.8 mils
2.9 mils
3.8 mils
Peak Density
6/mm2
17/mm2
7/mm2
Peak Count
49/cm
66/cm
57/cm
Characteristic
Yes No
% Change
Did re-blast decrease surface profile? X
-39%1 & -21%2
Did re-blast change the peak density?
+14%
Did re-blast change the peak count?
1Depth Micrometer 2Replica Tape

17. Test Results Variable No. 3
Initial Peak Density (Quadrant 1)
Peak Density Generated by Second Abrasive on Mill Scale (Quadrant 4)
Peak Density Generated by Re-Blast (Quadrant 2, 3)

18. Test Results Variable No. 4
Deep profile (e.g., 5-mil) with Steel Grit. Can it be reduced with a smaller abrasive (nozzle perpendicular)?
Technique
Attributes of Initial Blast on Mill Scale (Q1)
Attributes of Second Blast on Mill Scale (Q4)
Attributes of Re-Blast
(Q2 & Q3)
Depth Micrometer
7.0 mils
2.4 mils
3.1 mils
Replica Tape
5.9 mils
2.6 mils
3.7 mils
Peak Density
5/mm2
16/mm2
6/mm2
Peak Count
36/cm
77/cm
54/cm
Characteristic
Yes No
% Change
Did re-blast decrease surface profile? X
-56%1 & -37%2
Did re-blast change the peak density?
+17%
Did re-blast change the peak count?
+33%
1Depth Micrometer 2Replica Tape

19. Test Results Variable No. 4
Initial Peak Density (Quadrant 1)
Peak Density Generated by Second Abrasive on Mill Scale (Quadrant 4)
Peak Density Generated by Re-Blast (Quadrant 2, 3)

20. Test Results Variable No. 5
Deep profile (e.g., 5-mil) with Garnet. Can it be reduced with the same abrasive (nozzle parallel; ~30°)?
Technique
Attributes of Initial Blast on Mill Scale (Q3 & Q4)
Attributes of Re-Blast (Q1 & Q2)
Depth Micrometer
3.9 mils
5.6 mils
Replica Tape
4.4 mils
5.2 mils
Peak Density
5/mm2
4/mm2
Peak Count
58/cm
50/cm
Characteristic
Yes No
% Change
Did re-blast decrease surface profile? X
+30%1 & +15%2
Did re-blast change the peak density?
-20%
Did re-blast change the peak count?
-14%
1Depth Micrometer 2Replica Tape

21. Test Results Variable No. 5
Initial Peak Density (Quadrants 3 & 4)
Peak Density Generated by Re-Blast (Quadrants 1 & 2)

22. Test Results Variable No. 6
Deep profile (e.g., 5-mil) with Steel Grit. Can it be reduced with the same abrasive (nozzle parallel; ~30°)?
Technique
Attributes of Initial Blast on Mill Scale (Q3 & Q4)
Attributes of Re-Blast (Q1 & Q2)
Depth Micrometer
6.3 mils
6.8 mils
Replica Tape
6.1 mils
Peak Density
5/mm2
4/mm2
Peak Count
40/cm
37/cm
Characteristic
Yes No
% Change
Did re-blast decrease surface profile? X
+7%1 & -3%2
Did re-blast change the peak density?
-20%
Did re-blast change the peak count?
-8%
1Depth Micrometer 2Replica Tape

23. Test Results Variable No. 6
Initial Peak Density (Quadrants 3 & 4)
Peak Density Generated by Re-Blast (Quadrants 1 & 2)

24. Conclusions Test Variables 1 and 2:
Shallow surface profile can be increased with a larger abrasive
Shallow profile had no significant influence on:
New profile depth
Peak density or peak count

25. Conclusions Test Variables 3 and 4:
Deep surface profile can be reduced with a smaller abrasive (reduction of 1 mil to > 3mils)
Surface profile was deeper ( mils) than when smaller abrasive was used on mill scale
Re-blast increased peak count & peak density (verses deep profile)
Peak count & peak density both lower compared to use of “correct” (smaller) abrasive on mill scale

26. Conclusions Test Variables 5 and 6: Prevention:
Deep surface profile cannot be reduced using the same abrasive (parallel re-blast)
Re-blast reduced peak count & peak density
Prevention:
Prepare a test section (frequently required for cleanliness, but not roughness)
Particularly beneficial if pre-existing profile is revealed; or severe pitting

27. Conclusions
Many factors potentially influence surface profile depth and other associated surface characteristics1
Blast Nozzle Type & Size
No. of Blast Nozzles
Abrasive Manufacturer
Abrasive Type
Abrasive Hardness
Abrasive Shape
Abrasive Size
Blast hose length & diameter
Compressor size (CFM)
Air Pressure (PSI)
Type of surface
Hardness of surface
Factors change from project to project. Extra time spent at project start-up can significantly reduce re-blast costs and impacts on schedule.
1Reference SSPC-PA 17, Appendix B

28. 10/10/2019 2:52 AM
Too Deep or Too Shallow… Can Surface Profiles Be Changed by Additional Abrasive Blast Cleaning?
Questions
© 2007 Microsoft Corporation. All rights reserved. Microsoft, Windows, Windows Vista and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries.
The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.