UT thickness on Vinyl Ester lined tanks

Date 08-08-2004 04:20

Not exactly a CWI question. Along with my Weld Inspectior duties I also do tank inspections. ACCP UT II. We have a 0.375 in. thick tank with internal Flake Glass Reinforced Vinyl Ester coating system (3 coats totaling 50 mils)and coated externally with Semcrete No. 954 Epoxy (12 mils).
1. Is it possible that when we take thickness measurements from the outside of the tank that the sound wave propagates throught the Vinyl Ester coating and reflects back and is included in our thickness readings? We are using a simple StresTel T-Mike E thickness gage. We are getting readings in the mid 0.400's.
2. Do we have to much tank for our little thickness gage?
3. Do we need to invest in a gage with B-Scan or A-scan capabilities to separate the differ signals?

By thirdeye (***)

Date 08-08-2004 05:30

Hi dlmann,

Using conventional D-Meters, it is impossible to get accurate thickness measurments through coatings because the thickness of the coating becomes part of the thickness measurement. You are most likely getting your thickness reading error from the 12 mil thickness of the external coating. A simple mockup test panel will prove this out. Using a .375" steel plate, coat one side duplicating the exterior coating (material and thickness). On the opposite side, mask off one half of the surface and duplicate the internal coating (material and thickness) on the other half.

The method used to take thickness readings through coatings is called peak-to-peak or multiple echo. This technique uses at least 2 high amplitude multiple backwall echo signals and 2 gates to trigger the measurement. Typical transducers are single element delay line types. The contact face helps to dampen echoes from within the coating layer.

Most manufactures offer equipment that has this feature. Krautkramer's DMS (now discontinued) was a early machine that also had a A-scan display (for signal verification) and a data recorder.

By dlmann

Date 08-08-2004 19:58

Thirdeye: Thanks for taking the time to read and respond to my post. Since I began reading through the BBS sometime ago, I've read many of your replies and have come to the conclusion that you've been to more than one rodeo.

I remember in my Boiler Rat days at Longview Inspection using Panametrics DL36's (they have A-scan). Always had it set to read Peak-to-Peak. We used the DL36's for everything. I do not recall any experiences with lined tanks though, and outer coatings were always buffed off (if lead was not an issue).

I expected that it was probable that the outer coating was included in the reading, but because of the velocity change between the outer coating and the tank shell I thought it would be negligible (maybe plus or minus two or three mils). I may have been wrong thinking along those lines. I am taking you advice about making a mock-up.

By thirdeye (***)

Date 08-09-2004 03:48

dlmann,

A general rule of thumb in thickness gaging is: .001" of paint = .003" of steel. So you may have readings that are .036" high if the paint thickness is consistant. Let us know how your experiment comes out. A mockup (or a painted step wedge calibration block) is an inexpensive and practical way to go.