A286 Fastener Design

[jimbod20]

I’m hoping to find an MS9556 A286 fastener endurance limit to use for fatigue analysis. I find reference to numbers which vary significantly.
A simplified picture of the fastener installation/assembly is attached.

A simplified description is a stack of aluminum plates clamped with two MS9556-32 bolts.
o The minimum specified MS9556-32 RT ultimate tensile strength is 130 ksi.
o The minimum specified MS9556-32 RT .2% offset yield strength is 85 ksi.

I calculate an installed preload of ~1200 lbs per bolt at RT with an installed torque of 40 in-lbf.
o The installed torque range is 30-40 in-lbs.
o The thread tensile stress area is ~ .0200 in^2.
o Installed stress is ~ 60 ksi.

I calculate a hot operating temperature load of ~1900 lbs per bolt due to thermal growth.
o The calculated hot operating load is ~110% of yield strength.
o Operating stress is ~ 95 ksi.
o Should I try and lower this stress?
The bolts are subject to cyclic tensile loading only.
o Nominal load is ~1550 lbf. (77.5 ksi)
o Alternating load is +/- 350 lbf. (17.5 ksi)

I need to superimpose the nominal or mean and alternating stress on a goodman diagram to confirm I have adequate fatigue life. I think I’m ok on a Goodman diagram [URL unfurl="true"]https://res.cloudinary.com/engineering-com/image/upload/v1523397314/tips/cross-section_qsuilv.tiff[/url]with a UTS of 130 ksi and an endurance limit of 65 ksi. I’m just using 50% of the UTS. What should I be using for an endurance limit?

 

[rb1957]

with this loading is preload helping you ? the cyclic stress is from thermal loads, which just adds to preload. ok, you're getting some R effect but delta_stress is the thermal loading.

another day in paradise, or is paradise one day closer ?

 

[jimbod20]

I do have material certs for the bolts I have purchased and find the following.

The minimum required load at failure is 2600 lbs. (2600/.020 ~ 130 ksi).
Actual mechanical test results are all slightly greater than ~3850 lbs. (3850/.020 ~ 192.5 ksi).

I suspect the bolts I buy in the future will be similar to my first production buy, however, I'm using min specification strength for the analysis.

Comments?

 

[WKTaylor]

whoa cowboy...

Quote...

A simplified description is a stack of aluminum plates clamped with two MS9556-32 [A286] bolts.

o I calculate a hot operating temperature load of ~1900 lbs per bolt due to thermal growth.
o The calculated hot operating load is ~110% of yield strength.

What are Your trying to accomplish here?
Most structural aluminums are rated up-to-250F--350F service [or slightly higher, depending on alloy/temper] and have very large thermal expansion coefficients.
The A286 bolts are usually rated for up-to-1100F service and have very modest thermal expansion coefficients.

In this case, aluminum will expand substantially more than the bolts [same temp]... creating a net tensile stress increase on the bolt head/nut... while increasing clearance between the hole and the shank. Regardless, most high strength aluminum alloys degrade quickly above 400F.

IF, on the other hand, this assembly were to be hit with cryogenic temperatures [-250F] the aluminum would contract substantially more that the A286 bolt/nut and the aluminum would tighten on the bolt shank... but contract so that the net tensile pre-stress disappeared.

IF the thermal-cyclic range was [say] -250F to +350F... his would play havoc with Your dissimilar metal joint.

Some references may come in handy...

SAE AIR809 Metal Dimensional Change with Temperature
SAE ARP700 High Strength Elevated Temperature Bolting Practice
MMPDS-* Metallic Materials Properties Development and Standardization (MMPDS) or MIL-HDBK-5J METALLIC MATERIALS AND ELEMENTS FOR AEROSPACE VEHICLE STRUCTURES




Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

 

[SwinnyGG]

Is the aerospace industry still using Goodman diagrams for fatigue life prediction?

This is scary, as they are known to be inaccurate (often wildly so) and have been for a long time.

 

[jimbod20]

Hello,

I'll try again.
I should have been more specific with my questions. See below.

The preload on the bolts (~1200 lbs) at assembly is due to the installed/assembly torque (40 in-lbs). I'm assuming ~10-11% of the applied torque results in bolt preload. I lubricate the bolt threads when installed/assembled. I install a NAS62010L washer under the bolt head. I'm using shigley-mitchell, Bickford and websites referenced in other engineer-tips message threads to calculate the bolt load.

While installed in service the bolted assembly is just subject to temperature cycles. The in service temperature range is -65 deg-F to 300 deg-F. The change in bolt load is the result of the temperature change. I use typical thermal expansion coefficients for the steel bolts and aluminum plates. I quantify the equivalent joint stiffness and then calculate the corresponding change in bolt load due to the change in temperature.

I provide calculated numbers.
Room Temperature assembled: ~1200 lbs.
Hot 300 deg-F: ~1900 lbs.
Cold -65F: ~770 lbs.
The cold clamp load is adequate for assembly function.

I need to review the equivalent joint stiffness calculations. My numbers might be a little conservative resulting in a conservative cold to hot load range.

Questions/Comments/Please:
Do these calculated load numbers seem reasonable for an MS9556-32 fastener?
Is the calculated hot load excessive for an MS9556-32 fastener?

I was in a preliminary design review within the last several months with two global tier 1 aerospace suppliers and watched life limited safety critical parts analyzed in great detail to demonstrate/quantify life and watched simple fastener fatigue life justified with hand/spreadsheet calculations and a goodman diagram. These individuals posses a great deal more knowledge than I regarding this subject. I think simple is preferred if providing adequate fidelity. The PDR fastener loading is similar to mine.

I do find another e-tips message thread (32-421685) I think provides helpful discussion and reference to other methods to calculate/quantify fatigue life. That being said, the reference abstract denotes modified goodman UTS uncertainty. In general the thread discussion is helpful.

Questions/Comments/Please:
I recognize the Smith-Watson-Topper method is a reasonable choice.
I'm still struggling to find fatigue data I should use for an MS9556 bolt (A286 per AMS 5731) given the loading outlined above. I have looked at MMPDS, Mil handbook 5 and so on.

 

[jimbod20]

My Bad,

The reference thread is as follows.

https://www.eng-tips.com/viewthread.cfm?qid=421686

 

[WKTaylor]

OK... I think you have provided a clearer problem definition for this forum.

NOTE.
Your [steel/cadmium plated] washer PN is NOT NAS62010L... it is actually coded NAS620-10L.

CAUTION The OD of Your selected washer is smaller than the flange-size of the bolt head... reducing the bearing on the mating aluminum surfaces. Also the washer ID is a bit tight into the bolt fillet radius for my preference.

Sooooooo... what nut/washer are You using [on opposite side of the Assy?]? Nut tensile strength should match bolt tensile strength...



Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

 

[jimbod20]

I install helicoil thread inserts into the last aluminum plate. The thread insert is pn MS21209F1-15L. The bolts are threaded into the thread inserts.

 

[tbuelna]

Since you are installing the bolt into a self-locking helicoil insert, and you are relying on wrenching torque to control axial preload, then you need to account for prevailing torque of the insert locking feature. If you look at the procurement spec for the MS21209 .190-32 size insert, it lists a locking torque of 13 in-lbs max and 2 in-lbs min. In my experience, the locking element prevailing torque of new inserts tends to be close to the upper limit. With all metal inserts/locknuts, a deformed thread locking element prevailing torque value can be difficult to control precisely during manufacturing, especially with smaller thread sizes. So it is important to verify prevailing torque at every installation of your bolts, and adjust your final wrenching torque value accordingly.

 

[jimbod20]

Thank you.
The specified minimum and maximum torque values are in addition to running torque.

Running torque varies in accordance with insert variation.
Running torque varies with rate of fastener rotation.
Running torque decreases if the fastener is removed/re-installed etc.

I do believe we have reasonable control over assembly processes.