IRIS on Heat Exchanger tubes 2

[Rubje]

hi all,

I had two different contractors perform IRIS on one of my heat exchangers. Their data and findings are very different.

Any suggestions on how to move forward with this? our budget is limited so contracting a third technician may not be doable.

What questions should I ask them to ensure they were done properly?

Confirm the pull speed?

Water pressure?

any help would be greatly appreciated.

 

[0707]

A critical aspect of IRIS is ensuring that the mirror is at the center of the tube. An off-center ultrasonic pulse yields a distorted scan image because of the different ID and OD wall sound paths. That’s why IRIS kits are normaly equipped with centering devices helping operators keep the system centered.

Because IRIS is an ultrasonic technique, it requires a couplant. In this case, water. Tubes under test must therefore first be flooded to use this technique.

tubes need to be properly cleaned

more information in

https://www.eddyfi.com/en/technology/internal-rotary-inspection-system-iris

 

[EdStainless]

How many tubes have you pulled?
With any NDT the only way to verify is by cutting up a tube.
Selecting a reference standard that had discontinuities that are too different from the actual tube damage will make it nearly impossible to correctly size indications.
Both of these people should have supplied you with detailed procedures (signed off by a LVL III) that gives all of the details on the setup and the ref std.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[Rubje]

0707: That's a good point. is there a way to look at the data after the fact and tell if the mirror was off centered or was moved away from its starting point?

Regarding the water, will IRIS be able to pick up defects in the tube where the tubesheet is? I would think that region may not be full of water at that point and may not give accurate readings.

Thank you for the link. Is there a resource you recommend that contains more detailed information? In particular i'm looking for issues that can give an inaccurate reading.

EdStainless: i haven't pulled any but looks like i may have to pull one.

 

[0707]

Limitations of IRIS

IRIS is not suitable/reliable to detect small pittings and cracks

It is not or only hard to apply in confined space

Bended tubes (U-tubes) cannot be inspected within the bended area.

Tubes to be tested have to be perfectly clean to the bare metal in order to acquire adequate data.

Compared to techniques such as eddy current, the scanning speed is considerably slower. The pulling speed; 2 in/s

It is sensitive to ID / OD deposits and fins, which are not defects.

It has a hard time assessing through-wall defects because they look like missing data.

Deep defects lead to false calls, mainly in thin tubes.

 

[Rubje]

0707:

"Deep defects lead to false calls, mainly in thin tubes."
Can you elaborate on this point?

Also, any comments on my prior question regarding the section of tubes in the tubesheet?

Thanks again for all of your help

 

[EdStainless]

If the standard has reference indications that are a different size or shape than the actual indications then sizing is very problematic.
As tubes get thinner this becomes a bigger issue.
I have done NDT on tubes from 0.240" wall down to 0.018".
The reference std must be the same alloy and dimensions as the tubes being tested.
The suggested indications in ASME are basically suited for boiler tubes, not HX.


= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[Rubje]

Wow, this is news to me.

If a user has an old heat exchanger that has no prior inspections done then guessing the flaw sizes and shapes can be detrimental to the testing especially for thinner tubes.

I imagine there isn't much else to do in this situation, correct?

 

[EdStainless]

When you clean the tubes prior to inspection (they must be very clean) you can go in with a borescope and see what you are dealing with.
Knowing the metallurgy helps also.
For example in Stainless steel you are often looking either at small pits (usually <0.060" wide, and sometimes 0.010") or very fine cracks. Having these kinds of indications in the ref std can set you up for success.
In CS wider more open pits, or grooves would be more common.
In Cu alloys erosion is a common issue, with wide areas of wall thinning, but pitting is also possible.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[Dhurjati Sen]

You have the history card of the exchanger.

Pulling out a few tubes and corelating with the reports is an option, but then it seems you have time constraint, else you would have gone for retubing.

Practical solution is to accept the better report, junk the other one, plan for retubing / replacement at the earliest.

DHURJATI SEN
Kolkata, India

 

[EdStainless]

Unless one of the testers has been qualified on tubes with known indications in them then neither report can be trusted fully.
With all of the work involved in opening and cleaning a HX not pulling a tube is being cheap and lazy.
Many companies have saved damaged tubes over the years and built-up test bundles using tubes with known indications in them.
They use these for qualifying test procedures and technicians.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed