Ultimate abuse load - is a check at limit required?

[LtBuzzkill]

Hello.

I often have to size or analyze secondary structures to abuse loads, the requirement for which I always see written as an ultimate load (150 lb, 300 lb, etc.) applied at the most critical location. Although I've never been able to find it explicitly written in the structures manuals or in the requirements of our customers, the senior engineers in my office often require a check at limit load as well, such that the defined ultimate load is divided by 1.5 and the stress is compared to yield allowables. In cases where the critical failure mode is bending (and I find that it is quite often the case for the structures I’m checking), the limit load always determines the size of the structure because the ratio of ultimate to limit allowables is greater than 1.5 when plasticity is considered.

My thinking is that if it were intended that the structure not yield under abuse load as well as not break at ultimate load, then the abuse load would be defined as limit and it would be explicitly stated that an ultimate check at 1.5 times the limit load is required. That would leave no room for misinterpretation. As I understand it, that's the relationship between limit and ultimate flight load conditions. I think that same relation may be misapplied here.

So, in your experience, are abuse loads required to be checked at both limit and ultimate stress levels, whether or not that requirement is explicitly stated, or are our seniors oversizing structure? If anyone has insight into why abuse loads are defined as ultimate, and so often there is no yield requirement stated, it would be much appreciated. Thank you in advance for your replies.

LBK

 

[rb1957]

I've never heard of a proof check for "limit" abuse loads, but why not ? For most Al alloys it should be less critical than the usual ultimate check (fty > 2/3*ftu).

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

 

[LtBuzzkill]

Thank you for the response, RB, but that doesn't completely answer my question. I am specifically asking about bending in the plastic range. I can see for myself that multiplying the stress by 2/3 to check a fictitious "limit" condition won't fail the structure if the stress at ultimate load stays within the elastic range. However, 'ultimate' says to me that you're allowed to let the structure go plastic. This is when my bosses insist on a check for yielding at "limit" abuse load, although I do not ever see that condition defined or required in writing. I never see a requirement that ultimate stresses remain within the elastic range, either.

So here's my dilemma: when fty < 2/3 * fbu, then the structure can pass at ultimate load (required) but fail at limit load (not clear). I'm questioning whether or not this limit condition is required for abuse (not flight/ground/crash/etc) conditions and why. I am permanently overruled at work on this issue because of long-established precedent, so that's how I have to do things. However, I want to get perspective by asking how things are done outside my little corner of the world.

So I guess now my question is now two-fold:
(1) Are stresses from ultimate abuse load required to stay within the elastic range, and
(2) If not, does going plastic trigger a "limit abuse" load check?

Thanks,
LBK

 

[SWComposites]

No.
Yes.

Why would the end customer want a structure that could have permanent defomation from the abuse load?

 

[LtBuzzkill]

Hi, SW.

I understand that permanent deformation is undesirable for any structure. But then why write the load requirement as an ultimate condition, where by definition it is allowed to deform? Why not just write it as a limit load condition if that is what sizes the structure anyway? Is it generally expected that the engineer analyze to the written requirement (ultimate load), and then back into the real requirement (limit load)?

It's kind of like when someone asks me a question, hoping I'll meander into answering the question they really want answered, which is quite different. Then they get frustrated at me when I answer the question they asked rather than the question they wanted answered! Such is life.

LBK

 

[rb1957]

I agree with SW. It is generally understood that abuse loads are ultimate. Using plastic bending to pass them is reasonable. But then it naturally poses the next question about to onset of plasticity, and so a limit load check. This would be particularly relevant to a mechanism (is this what you have?), and particularly for flight controls. And particularly if the abuse load is not much higher than ultimate operating loads.

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

 

[Kwan]

The answer depends on your design criteria. So for example, "Part 25.305 (a) The structure must be able to support limit loads without detrimental permanent deformation. At any load up to limit loads, the deformation may not interfere with safe operation."

Thus, if your structure will cause detrimental permanent deformation you must do a limit check (ie: the deformation causes the elevator to jam). The limit check would then be flight limit loads (limit g, aero, etc).

 

[rb1957]

'cept in this case it's a limit abuse load ... which is a little bit odd. Where I can see the value is if the plasticity is not immediately apparent. If some subsurface element fails and the surface looks fine, then that is a problem.

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

 

[Stezza]

Don't forget that there is also a yield plastic bending allowable (this accounts for the fact that some thickness of materials from the outer fiber has to yield before a part permanently deforms). So for materials that have Ftu and Fty close together such as 7075 this means that the ultimate plastic bending allowable may be critical.
For example see the NASA Astronautic Structures Manual N76-76116 (available online) Figure B4.5.5.5-12 for 7075-T6 Extrusion.
Ftu = 75 ksi
Fty = 65 ksi
Plastic bending allowables for k=1.5 (rectangular section)
Fb_ult = 110 ksi -> 110/1.5 = 73.3 ksi
Fb-yeild = 78 ksi
Thus the part is critical for the ultimate plastic bending allowable.

 

[VEBill]

In general, Proof of Compliance can be accomplished by any of Test, Analysis, Inspection, or Demonstration.

These four categories and their names or definitions may vary slightly in different regulatory contexts.

I know nothing of this specific topic, but it seems like it might be a case where Test or Deonstration would not be the best approach to Proof of Compliance.

Note that these can be built up into little hierarchies, where an Analysis may need to include a bit of Testing on smaller bits.

Assembling it all into the Proof of Compliance then becomes Requirements Traceability. Break out the DOORS database.

I'm not sure it this helps. If not, disregard.

 

[LtBuzzkill]

Thank you for the responses, everyone. I have more to discuss on this and more questions to ask, and I have to do it without violating my company's social media policy and customer NDAs, so please bear with me.

To rb1957, what I am looking at is a mechanism. So permanent deformation could be detrimental to the operation of the device, obviously. All of the deformation requirements are specifically for operating conditions. There are requirements of no breakage at ultimate load and no permanent deformation at limit load, the ultimate factor being 1.5, also under operating conditions. However, the ultimate abuse load requirement that I speak of is specified for ground maintenance only, and the only criterion defined is an ultimate load. No yield criterion or limit factor applied there.

So, as far as I am concerned, the letter of the law is that there is no yield requirement for this abuse load. But does it make good engineering sense to check for that anyway? That is what I am trying to answer. One positive I can think of to doing it is that we would know at what load level the structure starts to yield. But I am uncomfortable with taking it upon myself to dictate what that load level should be without approval from the program. After all, there must be a reason that abuse load requirements are written as ultimate with no yield criterion defined. I don't ever want to go into a CDR and have to answer for why I added weight, cost, and additional design constraints (e.g. thicker metal gauges drive larger bend radii) when I didn't have to. If I ever do find myself in that situation, I want to be able to give a solid explanation.

Thank you, Stezza, for the tip on plastic bending yield strengths. I had no idea that yield strength would vary with shape factor. As far as I (and I'm sure most, if not all, the folks in my office) am concerned, Fty is just Fty, period. I'll have to research this, try to understand it, and see if it's something I can use in the future when these questions come up. I'd be a bit more comfortable if there were a graph in the NASA manual specifically for 7075-T6 clad sheet .040 or thicker, but I'll give it a whirl.

If anyone has more good insights into this issue, I'd love to read them. Cheers!

LBK

 

[rb1957]

how does your abuse load compare with your jammed cases ?

If you relied on plastic bending to clear your abuse load (a nominal load) then I think it's a sensible question, "how much load for onset of plasticity?". then compare that to typical operating loads.

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

 

[LtBuzzkill]

Thanks, RB. I think that is reasonable as well. Doesn't hurt to know what abuse load will permanently deform the structure, even if I'm not asked for it or given a load level to stay under. And it makes sense that the operating load criteria could be a reasonable guide as to whether or not what I get is acceptable. And thanks to Stezza's input, I'll try to make sure to know what is the real yield stress for sections in bending.

Thank you so much, all. I think I have a direction now. Of course, any more helpful comments are welcome!

LBK

 

[WKTaylor]

This topic has been addressed a few times before... with no clear answers/resolution.

Perhaps these might shed some old-light on the topic.

http://www.eng-tips.com/search.cfm?q=abuse+load&action=search

Abuse Loads

http://www.eng-tips.com/viewthread.cfm?qid=298478

Handling/Abuse Loads for Aircraft Interiors

http://www.eng-tips.com/viewthread.cfm?qid=260120

Interface loads for LRU

http://www.eng-tips.com/viewthread.cfm?qid=346839

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]