by Hennie de Wet
Recently, I had a discussion with the owner of a machine shop who claimed that liquid penetrant testing (PT) doesn’t work.
This machine shop is responsible for the refurbishment and repair of large hydraulic cylinders in the mining industry. Earlier that week, one particular cylinder came from the client with a leak.
After a chemical paint-strip, the cylinder was externally tested with PT. No indications or leaks were revealed.
On the hydraulic test bench under pressure, the cylinder leaked again. The exact area of the leak was identified and marked.
PT was performed again on the marked area, with the same results. No indications.
That’s when the owner called me to say, “Penetrant testing is rubbish and it doesn’t work.”
These cylinders operate under immense pressures and my first thought was that the crack would only open under pressure sufficient to cause the leak, but would close again as the pressure was released, making it impossible for the penetrant to enter the crack.
This is a regular occurrence on some aircraft wheels, so we are familiar with tight cracks and the limitations of PT.
Looking at the area of the cylinder they marked, I performed a quick surface clean to remove the old penetrant with solvent, gave it a few minutes to dry, and applied developer only.
After a few minutes, a thin wet line appeared in the developer. The hydraulic fluid trapped inside the crack bled into the developer.
Penetrant, which is a thin oil, simply cannot enter a crack filled with hydraulic fluid. The physics will not allow for that.
We performed a quick magnetic particle inspection (MPI) on the area and the crack was visible from a few meters away.
So, the owner now had to decide whether to perform a paint-strip then a deep, high-pressure, solvent clean on all of the cylinders prior to PT, or just perform MPI.
He bought a yoke later that day.
MPI and PT on the Same Test Surface
Sometimes we are asked to verify indications with another NDT method.
MPI indications may not be verifiable with PT, as the indications that may have been open to the surface are now filled with MPI contrast paint and carrier liquid. The PT results will not be reliable. Perform the PT first.
MPI after PT will not be a problem since the magnetic field penetrates the paraffin and any other contaminants that may be inside defects. The MPI results will be reliable when testing ferrous material. Perform the PT first.
Our job is to make the world a safer place, not to add to the chaos.
_______
Hennie de Wet is an ASNT and ACCP Level III in South Africa, hd.ndt.ct@outlook.com.
Photo courtesy of Hennie de Wet.