Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations GregLocock on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

How would you handle this repair disposition?

Status
Not open for further replies.

koopas

Aerospace
Aug 24, 2002
106
Hello again,

Here's the repair scenario involving a 2-inch long "T" section extrusion made from 7075-T6511, 0.100" thick.

The horizontal flanges of the "T" are attached to the airplane belly skin with two Hi-Loks on each side of the vertical flange. So, all four Hi-Loks are loaded in tension.

Now, the vertical leg of the "T" has two protruding head Titanium tension Hi-Loks loaded in shear that hold part of the lower wing-to-body fillet fairing. Both Hi-Lok holes possess the factory 2D+0.06 ED, and the holes are intact. However, corrosion was found AROUND one of holes on the extrusion's surface, between the hole and the leg's edge. In other words, corrosion was found in the edge distance "area" of one of the fastener holes. The local thickness remaining around that hole after blending out the corrosion is now 0.070".

Can you justify using the part as-is, or would you scrap the part and order a new one? What would be your methodology and substantiation in disposing of this discrepancy?

My main concern is that the bearing allowable of that Hi-Lok in the 7075 extrusion just plummeted since the bearing thickness was reduced by 30%. How can one still show this condition acceptable for strength? Or is it a lost cause?

Alex
 
Replies continue below

Recommended for you

koopas,

I tend to be inherently lazy, so this is my first approach at a repair of this type.

Since this is a small part, cheap, does not seem difficult to make or remove, it is my opinion to just scrap the part.

If shop or management balks at this, then you need to ensure that you can still transfer the required load into the part. This likely would require oversizing the adjacent fastener to recover the load capability lost due to the corrosion. Also, you need to do a net shear and net tension check to make sure that through oversizing and the corrosion removal, you still have the available strength in the part. Finally, fatgiue considerations must be made, as you are increasing the stress in the area of the blendout.

The time/cost spent for engineering, and rework of the corroded area, would likely exceed the time and cost to acquire and install a new part.

Best of luck,

jetmaker
 
You could also consider adding a strap onto whatever face of the leg isn't in contact with the fairing. I'm pretty sure I don't have a good mental picture of all that's going on, but sometimes you can add material this way. Is this T 2 inches long total, or is the T 2" wide and running continuously along the fairing?


Steven Fahey, CET
"Simplicate, and add more lightness" - Bill Stout
 
Sparweb,

The T is 2" long (i.e. you just cut a 2" long piece of extrusion), running fore and aft. Sorry, I am not very good in describing the problem (still working on using the proper jargon :)



Jetmaker,

I can show the decrease in bearing strength of that fastener hole due to the reduced thickeness around the hole. So I am able to quantify the reduction in load capability. In fact, if I use Fbru * d * t (which is questionable for a single shear joint), and if "t" is reduced by 30%, the bearing strength of that hole is also reduced by 30%.

In the end, I need to add the lost load capability to the other hole (Oversizing would be a good idea, as you mentioned)

Frankly, would the substantiation only involve using Fbru * d * t for the oversized hole, with an increased "d" of +1/64 (for 1st OS) ? Then, the difference between that hole's nominal bearing strength and the just calculated oversized hole bearing strength should match at least the 30% bearing strength lost at the other hole location due to reduced bearing thickness?

What if you run into ED problems due to the oversizing?

Hopefully, I make sense.

Thanks,
Alex
 
koopas,

That's the delema of oversizing. You risk ending up with short edge margin on the other hole, and more area out across the net section. It is a balancing act.

IMO, a 1st O/S will not make up the loss of 30% in bearing. However, I have not done any calcs at this time.

For the amount of time being expended on this, it would be more cost effective just to replace this simple part unless you are not able to get a replacement.

Regards,

jetmaker
 
Good day Jetmaker,

You're right. I couldn't regain the bearing strength lost with 1st or 2nd O/S. Going to the next size fastener did the trick though, as I regained a positive MS in bearing for the overall two-hole joint configuration. Luck was on my side, also, since I still had 2D+ ED with this next size fastener.

Now pretend the following scenario: you go with a 2nd O/S, calculate your new bearing strength with the increased hole diameter, and happily find that you regained the lost bearing strength.

However, you now find yourself with only 1.3ED. Since Fbru is a function of ED, your Fbru plummeted. In fact, if you recompute the bearing strength of that 2nd O/S hole with the lower Fbru, you may find yourself gaining only minimal bearing strength.

Question 1: do you usually ignore this reduction in Fbru due to the shorter ED?

Further assume that the slight bearing strength gain is enough to offset the loss of strength at the other hole location.

Now, you have a short ED condition to deal with.

Question 2: What would you do at this point? Would you do a shear-out/tear-out calculation and ensure that the shear-out load is higher than the recomputed bearing strength? If it is so, would you consider the issue resolved (i.e. the part can be reinstalled on the airplane as-is)?

Looking forward to your comments...

Have a good weekend ya'll!
Alex
 
koopas,

A1) you need to account for this reduced e/d on the oversized fastener, otherwise, what is to stop you from going way oversize. Not that most protruding head fastener bearing allowables are given for 1.7D edge margins. So if the initial analysis was don using the tables, then you have wiggle room if there was more than 2D before oversizing.

A2) You should always do a shear/tearout check, even if the bearing stress is acceptable. For example, very large bolt in a thin sheet with a 1.7D margin is likely going to be shear/tearout critical.

Now other checks are as follows: net tension and net shear. Make sure that as you oversize, blend, or do whatever that removes material from the cross section, you need to check the overall capability of the part. This is easily done since you can calculate the maximum joint load, and then check that against Ftu*Anet, or Fsu*Anet. After that, you need to do a compact bolt group analysis to make sure that you are not inducing any serious bending moments. In an ultimate analysis, bolt load is proportional to bolt shear area, and sometimes a weighting factor for material variances. So, if you oversize 1 fastener several sizes, this will shift the bolt group centroid so that it no longer aligns well with the part centroid, and possibly induce a moment about the part centroid. Finally, even modifying the part geometry by blending or trimming can cause shifts in part centroid such that an additional bending moment is induced. If it is a significant change, then it needs to be checked.

Hope this helps.

Have a great weekend all.

jetmaker
 
Good morning Jetmaker,

I appreciate your continued tutelage.

Regarding A1,

So you're saying that you would take into account the decrease in Fbru with a reduced ED? When you wrote "Not that most protruding head fastener bearing allowables are given for 1.7D edge margins." Did you mean to say that most protruding head bearing allowables are based on 1.7D ED? The "Not" at the beginning of your sentence threw me off...

Regarding A2,

I'd like to ask about the methodology regarding the post-oversizing checks. Please tell me if you agree:

(1). After oversizing, recompute the new bearing strength based on the decreased Fbru (due to shorter ED) and increased hole diameter.

(2). Compute the O/S fastener's new shear strength based on the O/S diameter.

(3). Perform a shear-out calculation based on the shorter ED and increased hole diameter.

(4). Perform a net tension calculation using Ftu*Anet (Anet is gross area minus the O/S hole area). Anet will be necessarily lower than the original Anet since oversizing the hole removed additional material area.

The lowest of the values computed in steps (1) through (4) is the new joint allowable of the O/S hole. Hopefully, the lowest value will be the one calculated in step (1), as the other calculations may not yield a susbtantial (or even positive) strength increase.

I have a 2 new questions:

1. you mention also performing a net shear calculation based on Fsu*Anet after oversizing. If one performs a net tension check based on Ftu*Anet, is the net shear calculation still required? In this particular repair, the net area is in tension rather than in shear so I am not sure if the net shear computation is justified.

2. As stated in the original post, there are two fastener holes on that vertical flange of the "T" extrusion section. One of them is the hole that has the material blendout around it. The other is the one we just oversized. When performing the net tension calculation in (4) for the O/S hole, what net area do you consider? Would you use Anet_o/s = (half the length of the flange - O/S hole diameter)* (original material thickness).

Thanks for your time,

Alex
 
Koopas,

Sorry for the typo. It was suppose to be "NOTE" not "NOT".

Basically you have the right idea with A2. It is prudent to check the whole joint capability, not just the single fastener that you oversized. For example, say you blend some thickness away at one fastener, and oversize others to make up the loss in load transfer ability. The net area for the joint needs to be looked at, not just at single fastener. I think you have the basic idea. Just wanted to expand a little.

New Questions:
A1) If the joint is subject to shear or shear and tension, then you need to check both Fsu and Ftu. If you feel the fitting is strictly a tension fitting, then a tension check only is sufficient.

A2) as stated at the beginning of my post, I think you should use the net area of the joint, so Anet = w*t - D_o/s*t - D_other hole*t_remaining - w_blend*t_blend. If you were to use the method you propose, which I honestly think is ok, you need to adjust the load on the O/S fastener to account for the fact that the larger diameter will carry a greater percentage of the joint load.

regards,

jetmaker
 
Good day Jetmaker,

Thanks for your insight. I want to make sure that I fully understand the implications of your latest reply.

Since you're stressing to use of the entire part's cross sectional area minus area of all fastener holes as the net area to calculate the net tension allowable, it seems logical that what you compute is the net tension allowable of the part, or this case, of the vertical flange.

As a result, instead of determining the joint allowable of each hole via calculations (1) through (3) outlined in my previous post, it would make more sense to compare the net tension allowable of the PART (computed in the above paragraph) to the SUM of all joint allowables in the part.

The joint allowables in this repair would be composed of the oversized hole and the undamaged hole. For each hole, one would perform calculations (1) through (3) [bearing, fastener shear, and shear-out] based on the correct ED and diameter of that hole. Of course, the O/S hole would have shorter ED but larger diameter while the undamaged hole would have longer ED but smaller diameter. The lesser of calculations (1) through (3) represents the joint allowable for that particular hole.

Then, the sum of the joint allowables of both holes is added to yield the vertical flange's "total" joint allowable, which is then compared to the net tension allowable which we've discussed. The lowest of the two is the overall strength of the "T" 's vertical flange.

Most importantly, to show the part good with the oversized hole, you have to show that the overall strength of the vertical flange AFTER oversizing is equal or greater fo the overall strength of the vertical flange BEFORE oversizing.

Does this process sound right? Or am I still fumbling?

Alex
 
koopas,

You are right on the money.

One thing I would add tho is a change in your terminology. For each fastener, you are calculating the FASTENER allowable. The summation of all FASTENER allowables yields a JOINT allowable. This is by no means an industry standard for terminology, but it might help to clearly separate what you are talking about.

Keep in mind, that the maximum load that the Tee fitting can handle is not just determined by the 2 fastener allowables in the Tee itself. It's capability is a function of all its loadpaths. For example, you might find that the attaching part has lower capability, or that the tension fitting effect of the 4 bolts to the lower skin is the weak point. The purpose of this note is that even if the O/S condition is worse than the before condition, look around at the adjacent structure to see if the other loadpaths are more limiting.

Good work.

jetmaker
 
Hi! you might try this. I take it this is not a pressurized fuse? a simple fix to your discrepency would be, if you have not extracted the surrounding fasteners.the remaining .070 thinkness and the 2d e.d. would be acceptible for fit form and function. just blending the corroded area and oversizing to the next 1/64 o/s hi-lok. but. 1 fix you might consider would be to fab a "L' shaped doubler to rest in the pocket of the blended fastener area. pick up the fastener on the horizonal part of "t" and upgrade to first o/s make the vertical leg of doubler "2inches long" and add 3 equal spaces and ad 3 .094 dia #30 ad material rivets to the vertical leg of "t" and faysurface seal with 1422 sealant and install. If you have already extracted the surrounding fastenrs if they where installed into interfearance fit holes you will need to go 1/64 o/s on these fasteners also.

Daniel Kilgore, C-17 Final Opps
Quality Assurance,
The Boeing Company,
President and CEO of
Lean Engineering Concepts "a new start up company , Lakewood, Ca. 90712
contact
 
Good morning Jetmaker,

Thank you for correcting my terminology.

I've thought about the other load paths where strength may be less than in the tee's vertical flange I've beat to death previously. Such an area, as you mentioned, could be where the horizontal flange of the "T" attaches with four Hi-Loks to the fuselage belly skin.

Is it possible to calculate that particular joint allowable, which encompasses the four Hi-Loks loaded in tension? The local fuselage skin thickness is quite thick, at 0.170".

Since we deal with joints loaded in shear exclusively, I am wondering if JOINT TENSILE allowables even exist. Any thoughts on this? I hope I am not opening another can of worms! :)

Alex
 
koopas,

The analysis of the 4 bolts in tension is another can of worms all together, and it is called Tension Clip Analysis. Do a search for tension clips, club foots, bathtub fittings, and Tee fittings. I would suggest you start another thread for the topic of Tension Fittings. It has already been somewhat discussed in these forums, so make sure you search here before posting.

Best of luck.



C17QA,

Don't remember, and too lazy to look back through all the text, but I think a 1st O/S fastener would not make up the difference for the material lost. However, I am willing to be corrected.

Later,

jetmaker
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor