Bolt hole eddy current inspections 1

[sdar79]

We have been transitioning a lot of our NDI procedures to remove fasteners and inspect with bolt hole eddy current. A lot of them are through stacked structure, say 3 layers of aluminum that are not taken apart. We are getting a lot of what I'll call false positives, crack indications that cannot be detected using other methods and when the structure is disassembled so each layer can be inspected individually, the flaw can no longer be found by bolt hole eddy current either. So I think we're getting a lot of noise due to the surface interfaces, and possibly due to burs and shavings inside the hole since they typically were not deburred during manufacture. According to our NDI group they can't improve the inspection to make the problem go away.

Does anyone have any insight or experience on this matter? Thanks in advance.

 

[dgapilot]

Sounds like the original manufacturing process was not very good. No deburr and no fod cleanup. A perfect storm for crack propagation.

 

[Sparweb]

Sounds like the holes were "match drilled" without a subsequent disassembly for deburring, cleaning, surface finishing or anything. Bad news, as DGApilot already said.
I think you will end up taking these joints apart no matter what. Bolt-hole EC inspection is not going to catch all the problems that they will have.


STF

 

[Tmoose]

If the inspection reveals manufacturing issues that should be corrected, would it be considered a success ?

 

[WKTaylor]

SDAR79...

Are You certain that the false indications have NOTHING to do with eddy current reacting to the fay-surface gap [laminate edge-gaps]?

NOTE. On certain old Boeing structure, tack-rivets were used to hold parts together... and then the primary fastener holes were drilled and filled with bolts. Most cases the added bolt holes were out-of alignment with the tack rivets [missing the center of the tack rivets] leaving fragments of tack-rivets along bolt-hole edges. This-one drove-us nut for awhile, until a fragment of a tack-rivet just fell out of one hole leaving a scalloped edge. A peek at the drawing confirmed that these rivets were used for assembly... and should have been destroyed when bole-holes drilled... supposedly. Obviously The EC probe was picking-up the edges of the tack rivet in the hole bores looking like major cracks on the scope!

Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true.
o For those who believe, no proof is required; for those who cannot believe, no proof is possible.
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion"]
o Learn the rules like a pro, so you can break them like an artist. [Picasso]

 

[tbuelna]

I believe fractures about the ID surface of fastener holes are not typically aligned the same as faying surfaces between the clamped structures. So if you are using the correct probe configuration, and have calibrated it to a reference standard appropriate for your particular application, then you should be able to minimize the potential for false indications. Most EC inspection procedures also list a maximum roughness for the surface being inspected (~125 microinch or so). So if permitted, it would help to prep any excessively rough hole surfaces using something like a light abrasive hone followed by a thorough cleaning to remove any debris. Of course this would require re-application of any corrosion protection required on the hole surface post inspection.

Here are a couple free online NASA documents discussing EC inspection that are worth reading:

http://hdl.handle.net/2060/20150020950

http://everyspec.com/NASA/NASA-NASA-STD/download.php?spec=NASA-STD-5009.016146.PDF

Here are a couple SAE standards for EC inspection of open holes in non-ferrous aircraft structures:

http://standards.sae.org/as4787/

http://standards.sae.org/arp4402/

If you don't already have copies of the two SAE standards, it's definitely worth spending $150 to buy them. They will likely provide help resolving your situation.

Best of luck to you.
Terry

 

[sdar79]

Yes, hole quality isn't always the best, but for a 30+ year old airframe it doesn't seem to have caused many problems. A couple other examples: I had one with 3 holes showing flawed, oversized the 3 and when NDI rechecked they marked 6 holes as flawed (some that were good just hours before). Or, after oversizing the crack indication has moved to another side of the hole, say 120 degrees off. It seems really finicky to me. I guess I'm just wondering if that's been everyone else's experience too.

 

[WKTaylor]

sdar79...

In addition to EC data suggested by Tbuelna...

The USAF NDI 'Bible' is T.O. 33B-1-1. The T.O. provides in-depth info on each of the primary NDI processes for technicians in Various chapters and is the foundation for all USAF aircraft NDI T.O.s. Eddy current is extensively covered. This T.O. is available for download at

http://www.tinker.af.mil/technicalorders/index.asp.

Also, some good info is provided in MIL-HDBK-728/2 EDDY CURRENT TESTING.

A few important questions.

What alloys are in the stack-up? Certain SCC/IGC/EXCO prone alloys are a pain: hidden defects can appear to spontaneously emerge/grow after minor blending/oversizing operations.

What level NDI tech is doing the job?

Is your NDI tech using a certified EC NDI test block 'standard' suitable for alloys being evaluated? NOTE: The 'USAF' EC test block standard in the 33B-1-1 is shown in Figure 4-39. Air Force General Purpose Eddy Current Standard. The test block is made with various precision-embedded defects so that tech can calibrate/practice their process prior to proceeding.

Are You able to get a rigid borescope into the hole to try visually detecting the 'new' EC defect indication(s), close-up??? Sometimes a close-up visual exam is worth the effort. IF a drilling/reaming linear/rifle scratch is evident, then You may have REALLY be experiencing workmanship issues. IF the defect is discolored, then IGC/SCC may be present... etc...

Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true.
o For those who believe, no proof is required; for those who cannot believe, no proof is possible.
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion"]
o Learn the rules like a pro, so you can break them like an artist. [Picasso]

 

[der8110]

What are the thicknesses of the 3 layers? Also what type of fasteners? Bolt hole EC is typically used for deep holes, i.e., each thickness is similar to or greater than the hole diameter. Also the "3 layers" statement piques my interest, understanding that may be just a 'for instance'. Again bolt hole EC is mainly intended for just one thickness, maybe two, where the parting line is well defined (such as on the flanges of cast housings).

I've seen all sorts of mischief when bolt hole EC is used on typical sheet metal construction (skin/doubler/stringer/frame flange), even worse when permanent fasteners are removed (rivets, HiShears, HiLoks) and non-destructive inspection becomes very destructive with walking drill bits, spun collars and so on. In those cases (with sheet metal thicknesses), surface probe eddy current on the outer layers is a better bet. Also consider the middle layer - is it fail-safe construction? Just a few additional considerations.

 

[tbuelna]

sdar,

What standard are you working to? What is the typical material thickness and hole diameter you are inspecting? What are you using for accept/reject criteria?

Cracks on the ID surface of bolt holes are usually assumed to run axially. Standard aerospace EC inspection practices are intended to detect flaws of around .1" or longer. If you have a stack of materials with a bolt hole thru them, cracks running axially may not continue from one part to the next. If only one part in the stack that has a crack is thinner than the rateable flaw length defined in your NDI procedure, what is your basis for rejection?

As I noted before, a proper aircraft EC inspection process should start with a written procedure. I'm pretty sure most of your issues might be resolved by making sure you are using the correct probe configuration for each inspection, and calibrating the probe to a reference standard designed to match each part feature being inspected.

 

[crackman]

sdar79

You got lots of great feedback and recommendations and this in no way trumps any of that. However, you might want to talk to the NDT folks. Normally they only perform inspections and stop there. I have overseen many bolt hole inspections on old aircraft and the first thing that pops up is the condition of the hole after removal of an old fastener which can lead to lots of falls calls. Normally, we allow the mechanics (not the NDT folks) to polish the hole prior to inspection but obviously without allowing any removal of material. This typically solves most of the false calls. Anyways, just my two cents worth. Good luck