Meaning of "fatigue load lbs" values provided in NAS spec (comparing fastener ductility/to

[bc1080]

Hello,

We are comparing several NAS bolt options as a replacement for an existing joint on a fairly short schedule and am looking into one we haven't used before. I noticed that on the NAS624-644 bolt spec sheet data table there is a header "fatigue load lbs +/-2%" with "high" and "low" columns of values under it for each fastener size. I have never seen this exact terminology used before and am trying to figure out how to interpret this data. First impression was that it was somehow related to the endurance limit or some pseudo-infinite life loading or performance in a testing sample set. But the high and low threw me, unless that is generic for "high/low cycles" or something like that. I am not aware if there is some sort of standardized bolt test for this that they might be referencing, the spec does mention fatigue test values. There is a huge range between high and low values, so I wouldn't think they would be acceptance values.
Ex: NAS638 (1-1/8")
high = 70,300, low = 7,030
They all seem to be 10x apart like that (equation based?)

Really at the heart of it we are just trying to qualitatively (numerically if possible, but very hard to find good, comparative values) compare the ductility/toughness/fatigue properties of A286 fasteners at 160ksi ultimate with Grade 8 type fasteners and the 180-200 ksi alloy steel bolts (like the NAS spec above). It's been somewhat hard to find consistent values for A286 in particular as I have found fairly wide variations of reported properties between difference references/vendors.

 

[SWComposites]

Those are likely the max and min load levels applied to the fatigue test.

 

[Sparweb]

These specifications (MS, AN, NAS, etc.) are procurement requirements that set a minimum standard. The load numbers mean "shall pass a given test". In the case of fatigue, this would be peak min & max loads that it must be cycled through. The figures you quote give a fatigue stress ratio of 0.1. If you look carefully, you'll probably find a "procurement spec" which will add much more detail. Many fastener spec's have static strength requirements too, sometimes even specifying both tension and shear strength.

People often mistake the requirements on the spec like this for "application guidelines" but that's not the case. If you do want to use this, you need to be very cautious. In fact, if you plan to refer to this data in any substantiation, it raises a red flag for me because this is a damn big bolt and likely to be a single-point load path for a major component (engine or landing gear). That kind of bolt should be installed with a huge pre-load. If you plan on your bolt's fatigue cycle to go as low as 7,000 pounds, which is a pitiful 5% of its strength, then you have drastically under-torqued the bolt. Ideally, the joint design does not relieve the bolt pre-load if the joint is loaded cyclically, and the pre-load should maintain a tension on the order of 30-50% of the bolt strength. Depends on the alloy, IIRC.

No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it.
STF

 

[bc1080]

Hey,
Thanks for answers so far. Sorry for the delay in response.
I think the comments are spot on, it's more of procurement/test value and it would be ill conceived to try to use these as a design value or something. The "low" number was throwing me, as mentioned, as any strength/preload value on a fastener that large should be much higher than 7000 lbs, so makes since it is more a parameter for the acceptance testing. Our preload is much higher, so we shouldn't be getting anywhere near that level with the operational loads. Just if you're curious, we actually have a multi-bolt application for these big guys (primarily tension).

Our problem now is that we are trying to determine the tensile yield stress (among other material property values) for this fastener. The spec allows 4140, 4340, and 8740 steel compositions, but no where in the fastener spec, steel specs, or heat treat specs can I find a solid reference that describes expected or min yield strength. I have found a yield value in the MMPDS alloy steel tables, but was told this doesn't provide an authoritative/traceable reference for the specific material(s) used in these bolts. Any other ideas on where that information might be documented?

 

[Sparweb]

Will you use the yield strength value to estimate joint static strength or are you going to use it to estimate/scale something more esoteric like a fatigue life or fracture model?
I would be satisfied that the lowest yield strength value in MMPDS for either 4130, 4340, or 8730 for any static analysis, but I don't have a feel for what that value does in a fatigue/fracture analysis. A very simple fatigue analysis wouldn't look at YTS at all.

If it is for LEFM analysis, these materials were tested for NASGRO: DOT/FAA/AR-05/15, Fatigue Crack Growth Database for Damage Tolerance Analysis

I do not believe the data in that report can be treated as "A-basis" or "B-basis" allowables, like they are presented in MMPDS. The difference between the "average" and the "95% threshold" may be the source of the advice you were told, but not sure what they actually meant.

No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it.
STF

 

[WKTaylor]

BC1080... Hmmmmmm...

"We are comparing several NAS bolt options as a replacement for an existing joint on a fairly short schedule..."

Proving more details on this joint would be helpful.

As I see it, so far, The 'existing joint' portion of this statement can go down 3-roads... [1] change in new production, due to non-availability/availability or strength/quality concerns; [2] remove-replace-existing-bolts modification [due to strength or other concerns]; or [3] replacement during a mod-repair. Also... What bolts are currently in-use... and/or that You also considered?

The 'fairly short schedule' portion of all this begs a question that haunts every project... are the parts-needed actually available in your time-frame from a certified source?




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]

 

[rb1957]

it sounds like this is a static allowables question, and the fatigue allowables is a "red herring" ?

I feel you're throwing a bunch of comments our way, and I'm not sure how they fit together ...

The bolt you're replacing is an NAS624-644 ?

The drawing allows different material fasteners ... 4130, 4340.

In my experience the difference in fty between different "flavours" of steel is due to the heat treat of the steel (and less to do with the specific formulation of the steel, eg 4130 and 4340 with 125ksi HT have fty = 100ksi (refer MMPDS-01, table 2.3.1.0 (g2).

if you're in so much hurry to solve this, how come a week between replies ?

 

[bc1080]

We would use the material yield data only for preload/strength ratio estimations.
It seems some confusion was created in that a material min yield value is normally provided for many commercial fasteners (ex. Grade 8), and that is not always the case for NAS fastener/materials, which defeats the analysis routine expectation of readily available material properties to work with. I realize in many high strength materials/fasteners the ultimate is the driving manufacturing and design value, and the yield is often a little less clear than with structural steels (usually found by offset methods).

This is a large bolted joint used for temporary assembly/fixturing purposes (not a permanent installation). The bolts currently in use for this joint are NAS9618 and one serious problem we have experienced with these is severe galling/seizing of some bolts and nuts (even taking many precautions) that have in a few cases required fastener destruction to remove hardware, which is unacceptable for the end user. This is likely aggravated by the current stainless on stainless combination of bolt and nut (NAS1805) and should in theory be improved by this alloy bolt. Additionally, the tensile yield values for the stainless bolt are relatively low (120ksi Fty is what is being used) and we are wanting to increase the preload in the joint to provide additional margin (it is a slip critical joint). We also considered NAS6718 and an enhanced-quality Grade 8 close equivalent (encountered too many hurdles with using non-NAS hardware). I think we are in your second category, in that we are replacing all the bolts as a part of a larger modification effort. The rush is just trying to select and order the fastener ASAP to minimize the potential delays, we aren't going to make changes based on fastener lead time once we know what we want.

 

[rb1957]

why is fatigue such a concern for a temporary fixture ?

deal with the biggest concern first ... galling due to stainless on stainless.

you're using NAS9618 now. these are 160ksi A286 bolts. I'd look for a Titanium alternative.

I wonder how galling sets up in a temporary fixture ??

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

 

[bc1080]

Missed rb1957's last response. I apologize for the confusion, the question/problem did shift somewhat after the first post. Initially the problem was merely trying to understand the "fatigue load" columns in the NAS spec, the issue morphed into determining where to find reference-able material properties for the bolt alloy(s), in particular the yield strength. There is a bit of a disconnect for some users as the tension/preload recommendations are typically based on a percentage of fastener yield but a min/max yield data not being directly provided for the fastener material in the fastener's spec or the pertinent steel specs. I had always assumed the MMPDS data was acceptable for this purpose, but it caused a little stir. Initially there were some diligence questions regarding fatigue/fracture/ductility properties of the new fastener vs. previous, but you are correct that with an extreme low number of cycles we are not really concerned (if we eventually decide to reuse bolts it will get a harder look).

We think the thread galling has been primarily due to bolt condition and installation factors: high torques, dirt/contaminates in threads, nicks and thread damage that was not found/dressed, rough installation techniques creating thread damage (use of impact tools or forcing bolts through tight holes). Seems like once there is damage in the threads it kind of snowballs. Lots of damage was observed in the bolts where they engaged the deformed thread region of the nut. Then they tried to remove the bolts with impacts and the next thing we knew is that some had chewed up threads and a few are stuck. I'd concur that Titanium (or Inconel if needed) could be good alternate materials.

 

[WKTaylor]

RE Galling... this is a problem with high nickel alloy Bolts and nuts mated together with high installation torque... dry.

This is easily solved by installing nut on bolts threads wet with an antiseize compound, such MIL-PRF-907 or MIL-PRF-83483, etc...

A product I have had good experience with [Nickel alloy 718] is Bostik Pure Nickel 'Never-Seez' ... or any one of the Never-Seez variants.

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]

 

[Sparweb]

...or a zillion variants of LPS dry film lubricant. Frequently seen in toolboxes around here.

No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it.
STF

 

[rb1957]

so our "recommendation" is not to change the bolts, but to install the nut on clean thread with anti-seize.

As for the preload discussion, most companies have design guidelines for installing in a shear application (low preload) or a tension application (high preload). If preload is That significant for the joint performance then I'd prefer to see specific torque called out on the drwg.

Is this a build, or a transport, fixture ?

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

 

[bc1080]

Hi again,

We had been using MIL-PRF-46010 heat-cured DFL on the fasteners before. I think much of the problem being the lack of "clean threads", since that was almost certainly not the case previously. We are calling out the torque on the installation drawings, in addition to cleanliness requirements and many other installation notes/procedures being that it is a slip-critical joint.

The original bolts were selected really only due to known availability and department familiarity. We didn't have a particular reason to use them (A286 is usually selected for permanent fasteners due to SCC resistance), so we figured it might be prudent to consider additional fasteners that might be better suited.

It's for a transportation and installation fixture, basically there is a very odd shaped/sized (very heavy) piece of machinery that has to be moved to the use site and installed. However, the designers did not include specific handling provisions in order to save weight/complexity, so we were boxed in to making do with the interfaces that already existed. This limited us to certain size/number/types of connections that we had to make work, which has made the fixture fairly intricate.

 

[WKTaylor]

IF I recall correctly, MIL-PRF-46010 SFL was never intended for high pressure lube in threading application... except as a 'back-up'... maybe...

More appropriate SFL for CRES fasteners/threads...

AS1701 Class III - Lubricant, Solid Film
AMS3090 - Lubricant, Dry Film, Heat Cured, For Fasteners Polysulfide Sealant Compatible

Or... just use a good antiseize thread lubricant compatible with Your structural materials.

Or... Cadmium plating on fasteners threads [nuts/bolts] with a light oil is also a 'workable' choice. Cadmium has high lubricity and is an natural 'antiseize' finish.

Or... maybe... Another possibility is silver-plated nuts. Silver has high-pressure lubricity and is an natural 'antiseize'... as well-as a high temp... finish. Works well on bare CRES threads parts.

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]