Magnetic Flux Inspection: The Bottom of a tank floor tell you a lot with an MFL scanner.

by John Taylor

When it comes to insuring the safety of any storage tank, no news is not necessarily good news.  The problem with storage tanks is that even when everything looks good on the outside, inside the tank can be a ticking time bomb, where an unseen defect or spot of corrosion can lead to a catastrophic and costly failure.  That's why magnetic flux testing leakage (MFL) testing is vital to the continued longevity of storage tanks across the nation.

Most out of service inspections use magnetic flux leakage (MFL) as a great detector for locating underside floor corrosion. This is often referred to as soil side corrosion caused by the contact of the steel floor with the moisture and chemical make up of the soils underneath the tank. MFL however cannot give an accurate assessment of the remaining wall thickness due to some of the limitations imposed by the inspection environment. The only reliable and accurate quantitative assessment must be obtained by using ultrasonic thickness techniques. These techniques include appropriate operator training and not just the ability to take digital thickness measurements.

This discussion concerns the added requirements for the training and testing of NDT personnel who inspect storage tank floors during an API 653 out of service inspection. The America Petroleum Institute added Appendix G to the 653 code in 2003 to describe the precise requirements for this training and testing. It describes the number of training hours, the types of test questions and even details the construction of a tank floor mock-up with manufactured defects.

Magnetic Flux Inspection Picture 1: MFL (Magnetic Flux Leakage) Scanner.  Model: MFE 2412 MK2

It is essential that a quality A-scan be used to assess any corrosion detected. This equipment must be capable of determining location, size and the accurate measurement of remaining floor thickness. A 5 Megahertz dual crystal focused contact transducer of either .375" or .500" in diameter should be used. Either water or a gel type couplant should be used. It is essential that the amplitude of the signal from the corroded back wall must be raised to at least the same screen height as the calibration reflector. While it is true that a reflector from a straight back wall will give a clean signal the same is not true for a corroded surface. Since there is an irregular profile the energy being reflected back will be wider at the base and have much lower amplitude. In this case the operator's experience in adjusting the ultrasonic equipment is essential. This is the reason that code book now requires the advanced training for ultrasonic prove-up operators.

Here is a summary of the requirements now in place for API 653 tank bottom examiners.

1. A written procedure that address the essential and non-essential variables for a tank bottom inspection.

2. A record of the qualification test for a specific scanning operator.

3. A record of the qualification test for a tank bottom procedure.

4. Each scanning operator shall receive 40 hours of training. The training will include NDT principles and scanning equipment operation and calibration.

5. Specific dimensions of the tank bottom sample (70 square feet) and a minimum number and size of pits to be located during the operator testing.

Magnetic Flux Inspection Picture 2: Tank floor that been scanned and mark. Ready to be A-Scan

Recently we did a tank inspection at na air force base using MFL equipment.  The task, while routine involved the inspection of a tank contining JP-8 (Jet fuel). Everything looked good on the outside and on the inside.  Visually the tank would pass muster, sporting good plate weld lines, no obvious signs of corrosion, even a fresh coat of paint.  After our visual inspection, we took some thickness readings around the corner of the plates and in the middle in order to get an average.   Now the funs begins with the MFE.  As soon as I started to scan the surface of the plate I noticed that the MFE scanner’s screen was lighting up like a Christmas tree.  Usually we do a cross scan of the area so we can narrow down the area so it’s more accurate.  Once any defective area is found and marked, a technician will do an A-scan so we can get the lowest thickness reading of the area and note it in our report. Surprisingly all the areas that I inspected had a reading lower than the average. This meant that the tank was far from safe to refill with JP-8.  All I could say to myself as I climbed back out of the tank was, "Never judge a tank by its cover."

John Taylor is president of Nova Data Testing, a leading Magnetic Flux Inspection contractor.  For all your magnetic flux inspection needs go to http://magneticfluxinspection.com/