Predicting Tensile Strength 1

[Danny Park]

Our company manufactures fuse pins that go into 777X Boeing commercial airplane. Following the ASTM E 8 procedures and Boeing's BAC specifications, the tensile test and double shear test are run.
In order to predict the durability, crack growth, and fracture toughness,Boeing wants to know how various properties of the fuse pin can be correlated to predict the Ultimate Tensile Strength. In summary, we need to provide Boeing with a model that predicts the end properties of Tensile strength with following inputs:
(a) shear strength ----> = π/4 * (O.D. - I.D.)2
(b) Inner Diameter (I.D.)
(c) Heat treat temperature ---> usually 1100 ±90 °F
(d) Copper weight & ----> 3.00% ~ 3.75%
The fuse pins for this model are made of 15-5ph Stainless Steel. I tried to search for direct relationship between the transverse tensile strength and the shear strength, and the only valid answer I could find was about Von Mises Criterion. My data do meet the apporximation of .577UTS = Shear strength. However, I am not certain what will happen when heat treat temperature and copper composition variation come into affection.
Thank you.

 

[Sparweb]

Hi
First off, welcome to Eng-Tips.

My understanding of your question is that you are speaking of the pins that mount the engines beneath the wings.
Not familiar with the model 777 specifically, but the 747 made the news a few times when these pins failed when they shouldn't have.
It might help if you tell us about your role in this manufacturing operation. Are you running a test lab, or are you the top dog in quality control?

Reading the spreadsheet you attached, I notice a few things. It seems (by reference to the ASTM E8 tests) that you are tensile testing specimens of the material.
Seeing that your specimens are cylindrical, and roughly the same OD as the engine mount pins should be, it appears you are tensile testing the actual production pins themselves, not coupons or samples of raw material.
The table does not specify which values are calculated, which values are test results, which values are "book" values from a material supplier, or derived from, say, correlation of the Rockwell C hardness with typical tensile strength. If you could clarify the meaning of the data in each column, I would be able to know what is being presented. It would also help to explain the source of the curve-fit relationships that are shown in the far right columns. The average of seven invalid curve-fits does not make a valid result, if you know what I mean.

Another thing I can't help but notice in the spreadsheet is that the OD tolerance is +/- 0.00010 inch (as expected) but the ID tolerance is hit and miss with about 0.050 inch error.
Are you roughly boring out the ID of the pins for some reason? Are they supplied to the testing lab that way? There may be a good reason for this but it's distracting my attention from more important parts of your question!

One last thing; while the title of your post reads "predicting tensile strength", the test procedure you list is ASTM E8, which is a Tensile test. You have already measured the tensile strength of the coupon. Therefore, no prediction of tensile strength is necessary.

Are you perhaps trying to "Predict Shear Strength"? These pins are loaded in shear, aren't they?


STF

 

[SWComposites]

What I suspect Boeing wants (expects) is that you run a bunch of tensile and double shear tests on pins with varying ID, heat treat, composition and develop the relationship to predict tensile and shear strength as a function of those parameters.

Actually I am somewhat surprised that Boeing has not already done this.

Anyway, why don't you ask Boeing for clarification on what is needed?

 

[metengr]

The approximation of UTS to ultimate shear strength can vary for materials from 0.40 to 0.60 of UTS. Where are the YS values from the tensile test in the spreadsheet? A better approximation is to determine the yield shear strength, which is 0.57 x tensile YS. I would want to know when the pin yields before fracture.

Last comment, Boeing should be evaluating this information versus relying on a vendor for this? Amazing.

 

[EdStainless]

Published data for 15-5 shows shear/UTS ratios of .62-.63
Are you annealing?
Are you sure that material was cooled fast enough from the mill anneal?
I would take parts, age them, re-anneal, cool, and re-age. This assures that the HT is correct and it offers grain refinement. (the material becomes finer grained and tougher with re-HT)
You can do some sensitivity analysis with final ageing temp, just run some trials at +and- 25F.
And look at Yield, variation there can have an impact.
I have never seen a reliable way to predict strength based on chemistry.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[israelkk]

What puzzles me it that you should have the spec. the 15-5PH is purchased tO. The spec gives the minimum UTS and Yield. You should do your calculations based on these values. The tests Boeing asks for to my best 40 years of experience in the area is to see that the pins material properties were not altered due to manufacturing uncontrolled processes. To be able to give reliable results you have to know the actual UTS of each batch of material based on the material manufacturer C.O.T or/ndn independent lab to verify the C.O.T data (on standard tensile test samples. Only then and knowing exact OD an ID of each pin you test you can trust the shear and tensile results and compare to the calculation.

 

[Screwman1]

I think back about 30 years or so to the DC-10 that lost an engine on takeoff from ORD because the fuse pin failed and they lost everyone on board. These are one of the most critical components on an aircraft so I am terrified that the vendor for them is asking an internet group how to make them.

 

[Danny Park]

Thank you for all the advice. I am surprised to see so many people taking notice.
Yes, I am surprised that Boeing does not have set rules regarding Fuse Pin issues, which has been manufactured for decades.
I am fairly new to this industry (just received my B.S. last June), and since I am the only material engineer among other mechanical engineers, I was given this duty (exciting!)
I apologize for not being able to show all the detailed data. I have a spreadsheet that demonstrates far more details and data than the one I posted, but I am afraid that revealing every data will get me into trouble

SparWeb: Thank you! Exciting to find out such helpful website with great forum. We do test some coupons, but for this specific job, we are testing the fuse pins themselves. However, the tensile tests are mostly done through outside testing vendors. We do bore and hone I.D. accordingly to Boeing's given drawings.
We are trying to predict tensile strength at this point.

SW Composites, metengr, israelkk, Screwmann1: Yes, this issue is vague to me as well.. I can come up with some kinds of relationship between all these variables and tensile strength through JMP analysis' linear regression analysis or Gaussian relationship. However, I was hoping to have a valid supporting fact of how these variables directly influence the tensile strength variation.

EdStainless: Yes, we are annealing the fuse pins through outside vendors. I also wonder what copper composition can do to tensile strength; my only knowledge is that copper composition in stainless steels enhances corrosion-resistance and promotes austenitic microstructure. May I know where I can find the published data regarding 15-5ph shear/UTS ratio?

 

[metengr]

Here is a publication. Look at page 3

http://www.dtic.mil/dtic/tr/fulltext/u2/755408.pdf

 

[3DDave]

Screwman1,

The most noteworthy crash of a DC-10, Flight 191, at Chicago was initiated by a structural failure of the wing pylon mount following a failure to follow the manufacturer's defined engine service procedure.

I recall reading the testimony at the time where one of the participants in servicing the engine said they heard a sound like a gunshot, but close inspection did not reveal any damage. In truth, a small crack in the rear pylon mount was created, but with the load of the engine removed the crack closed up and would have not been easily visible. 8 weeks later the crack reached a critical stage.

Sadly, in spite of all the remaining damage done, the loss was survivable, but there was no experience the pilots had that would have prepared them for the combination of failures they were faced with and insufficient time to sort it out.

 

[EdStainless]

Danny,
There are still the books around published by Armco The guys that did all of the alloy development work) back in the day. I have the one for 17-7/15-7, and a partial scan of the 17-4 ones. These books run 50-80 pages long. There are also data sheets that are 6-10 pages that have very good summaries in them, that is where I found the shear/UTS ratio data.
These are not austenitic alloys, they are semi-martinsitic even when annealed. The role of Cu here is to form the precipitation hardening phase. Some of the other PH alloys use Ti as the PH modifier.
15-5 is a modification of 17-4, it provides good properties in thicker sections, has better transverse toughness, and is almost always provided as remelted (higher cleanliness) material. The next step in the progression is 13-8 SuperTough, it is VIM-VAR melted and has incredible toughness.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[israelkk]

I assume Boeing has a minimum shear load the pins has to stand. Therefore, you have to show that at minimum material condition (smallest OD with Largest ID) the pin will hold at the minimum UTS of the source material at the required heat treatment given by the material spec. Statistics to my knowledge is irrelevant. Therefore, when you test your actual pins and the actual UTS based on the C.O.T or the lab results on standard tensile test samples you have to normalize the results to the minimum UTS that the spec allows. For example: If the spec minimum UTS is smaller by 10% than the actual pins you test's UTS, than the shear results of your test should be reduced by 10% and compared to the minimum shear Boeing specify.