Babbit disbonding

[tidraie]

A manufacturer asked me to inspect babbit disbonding on a cast iron sleeve. After several attempts to get a back wall deflection, i informed my client that this type of inspection (ut)was not possible.
Another technician inspected this part and called a total 100% disbonding.
A third inspector called a 30% disbonding.

I'm amazed to see that two ut tech said that it was possible and there are two totally different calls.

Both tech used a babbit step block for calibrating the instrument.

The cast iron is of irregular shape and contaminated with oil. The babbit is uneven with at least two thickness.

My question is how can one test for disbonding with only a babbit test piece?

 

[Corgas]

I'd start by asking both UT techs if they are certified according to ASNT (at least level II).

 

[arunmrao]

Corgas, I have very little confidence in certification. Recently I had a part just a simple cast cylinder. UT was clear (Level 1),while the RT report mentioned Shrinkage defects ( beyond repair). I was very confident of the part, I sectioned the part in various places as per the RT film and no defect was visible. I lost the part and time,who pays?

 

[tidraie]

I'm a certified NDE senior technician in all five methods(MT,PT,UT,RT and EC) to CGSB certifications which is equivalent to ASNT-TC-1A level 3(with 30 years of experience) and CWB (AWS) level III welding inspector.

I refused to carry out a ut inspection simply because of no back wall deflections was displayed. What i would like to know is it possible to verify disbonding of babbit/cast iron without a back wall deflection ?

If so, can anyone tell me how it's done.

 

[ndeguy]

tidraie

Go to fusionbabitting's website or fusionbabbitt on YouTube. There is a video of a UT bond inspection being performed and from what I can see of the screen most of the time there is no backwall echo (or bondline echo, either) visible. presumably the acoustic impedances of the materials are sufficiently similar to cause no bondline reflection.

If the relative difference in acoustic impedances of the bond material and the backing is sufficient, a (calculable) percentage of the incident sound energy will be reflected by the interface with the remainder of the incident sound energy being transmitted across the interface. I can imagine the 1st technician (100% failure) mistook this returned energy for disbonding and therefore called it 100% disbond. Perhaps the 2nd tech knew what he was doing and set his bond zone reflection from a visible good bond to a certain amplitude and then looked for an increase in that amplitude and those areas demonstrating an amplitude increase he called 30% disbond.

Was it the lack of BWE which caused you to say that UT was not feasible on this item, or the facts that "The cast iron is of irregular shape and contaminated with oil. The babbit is uneven with at least two thickness".

To answer your question "how can one test for disbonding with only a babbit test piece?" if the UT timebase is calibrated for x mm range of Babbitt material where x is greater than the Babbitt material thickness and the gain is increased until grass is say 5% then the test can proceed. If the materials' acoustic impedances are significantly different then a bobdline indication will show which can be monitored for amplitude increases.

 

[tidraie]

Just checked out on you tube and you can clearly see on the Panametrics Epooch 3 screen the IP (main bang) and the back wall deflection I presume. Note that there is excessive gain,range control to narrow,no first gate showing and the second gate at babbit interface is not in place.

The relative difference in acoustic impedance has crossed my mind during my inspection but there is to many unknown factors to take into consideration.

1- total back wall deflection loss due to large grain size,heavy oil contamination and irregular thickness and angle of the cast iron.

2- Cast iron sleeve has vee noches to retain babbit.

3- Babbit material is of at least two thicknesses 200 th and 350 th.

4- Angle beam spread in babbit material makes it impossible to clearly identify retainner's(cast iron) location.

Now for the change in amplitude that is a tough one to understand. At 350 th in thickness you can raise the amplitude to 80% FSH. Once you have locked the gain control and you scan over the 200 th material, it is quite obvious that the amplitude will rise to 100% and more.

This is where I have a problem in understanding this method.
But on even flat sufaces I think that your sugestion might just work.

Your comments are quite welcome.