MIL-HDBK-5/MMPDS ISSUE?!?!#&@# 1

[WKTaylor]

Folks... I need Your comments… especially You stress guys and DERs…

I have worked too many places and am now faced with a dilemma I never expected: Interpretation at to exactly what “A” and “B” basis allowables mean… and when/how it is acceptable to use them.

I always used “A” basis values whenever using “specification” metallic materials. NOTE: “spec minimums” invariably match HDBK “A” basis allowables. Note: use of “A” basis allowable-values seems very reasonable assuming relatively “loose” [typical] machining tolerances/finishes… and/or when “standard” sheet metal fabrication/forming is specified.

I reserved the use of “B” basis values when I had the “luxury” of lab-testing [mechanical, chemical] certifications for the EXACT piece of material to be used… and/or for a piece of material that was traceable to a lab-tested/certified lot of material. NOTE: when using “B” Basis allowables, I always used/specified VERY strict manufacturing tolerances and processes… and then analyzed to minimum dimensions/thicknesses.

From MMPDS para 1.4.1.1 regarding “B” basis materials….

“Use of B-Values — The use of B-basis design properties is permitted in design by the Air Force, the Army, the Navy, and the Federal Aviation Administration, subject to certain limitations specified by each agency. Reference should be made to specific requirements of the applicable agency before using B-values in design.”

My dilemma: I got a range of answers from (4) stress analysts on this issue, roughly paraphrased as follows:

I only used “A” basis allowables, because “B” basis values have an unacceptable level of uncertainty and/or risk... and I “assume” worst case manufacturing tolerances/practices will creep in. Besides You can’t get [specify] “genuine “B” basis material.

I always use “A” basis allowables… rarely “B” basis without lab certs.

I always use “A” basis allowables for primary single/critical load-path parts [lugs, spar-caps, critical fittings, etc]: and always use “B” basis values for secondary or redundant/multi-load-path structure.

I always use “B” basis allowables, because the material-vendors are capable of making the raw material to the “higher level of quality” … and why would they [“B” values”] be in the handbooks.

OK… what are Your thoughts on this issue [maybe we will also add “S” basis allowables to discussion, later]?????????????????


Regards, Wil Taylor

 

[SWComposites]

Wil - Well, I work mostly with composites where the situation is even more confused, and my experience is mostly with FAA certified structure, but in general for metal structure I have used the approach of stress analyst #3: A-basis for critical and/or single load path structure, B-basis for secondary and/or multi-load path strcture. The difficulty always is deciding when a structure is really multi-load path. The approach of analyst #4 is very naive and unconservative - this person has obviously never seen allowables datasets, material batch data nor any other real world data, and has the delusion that a B-value implies a higher level of quality - pure rubbish! And since you mentioned it, S-values are not in any way basis values or allowables but simply a spec minimum value for which there may be little or no data backing it up and for which there is NO statistical assurance. My experience is that most stress analysts do not understand the statistical assumptions behind allowable strength values, and many are shocked to find out that a B-basis value means that 10% of samples can have a strength less than the B-value. Many aso do not understand that a basis value is not a fixed value but is a function of (among other things) the number of samples used to calculate the value.
Regards,
Steve W.

 

[i278]

Wil,

SOP at the office where I worked (mod/repair cert.) was to always use B-values to calculate strength of existing structure, and always use A-values for our design work. So, I'd sit in the analyst #2 camp--I've just never had satisfactory lab certs at my disposal.

Regards

 

[rb1957]

wil,

i've always used A-basis, and i've always thought the A-basis/B-basis debate as hair splitting, i mean B-basis is only 5% better than A-basis typically and i'd keep that in my back pocket 'cause i'm sure i don't know my loads to that accuracy.

 

[RPstress]

SWComposites has the right of it for airframe design. S-basis can be relatively safe for preliminary sizing, but approaching certification the authorities' (basically FAA or JAA) representative should be demanding that testing be scheduled to generate A- or B-basis allowables as appropriate. The final allowable, even A-basis, almost always comes out higher than the S-basis. The material supplier usually goes to some pains to set an S-basis value that they can achieve fairly easily.

Increasingly I've seen a requirement to use a B-basis compression modulus for buckling calcs.

 

[737eng]

From FAR 25.613:
(b) Material design values must be chosen to minimize the probability of structural failures due to material variability. Except as provided in paragraphs (e) and (f) of this section, compliance must be shown by selecting material design values which assure material strength with the following probability:

(1) Where applied loads are eventually distributed through a single member within an assembly, the failure of which would result in loss of structural integrity of the component, 99 percent probability with 95 percent confidence.

(2) For redundant structure, in which the failure of individual elements would result in applied loads being safely distributed to other load carrying members, 90 percent probability with 95 percent confidence.


From MMPDS-01 Appendix A.3
A-Basis.—The lower of either a statistically calculated number, or the specification minimum (S-basis).
The statistically calculated number indicates that at least 99 percent of the population of values is expected
to equal or exceed the A-basis mechanical design property, with a confidence of 95 percent.

B-Basis.—At least 90 percent of the population of values is expected to equal or exceed the B-basis
mechanical property allowable, with a confidence of 95 percent.

I use A-basis if I have a single load patch structure (i.e. lug, joint w/ less than four fasteners...)

I use B-basis for mult-load path structure:
frames+stringers+skin (if one fails, the other will carry the load)
joints with four or greater fasteners

I do know other DER's who strictly use A-basis.

Another note, is if 99" pop is req'd and you are using an allowable that lists both A and S basis, you should always compare "A" and "S" basis to see which is more critical and utilize the more critical basis

 

[WKTaylor]

Wow...

Talk about differing approaches... You-all [or someone You know] seem to align with everyone on my survey and possibly more… except for the one who strictly goes with “B” basis.

737eng… I see where You get your rationale basis for the “mixed approach” comes from. My problem is that I work military programs… which do not conform to FARs… and seems to have less-clear guidance. NOTE: this data may lie within a document reference I just discovered [AFSC DH 1-2]… but, then, it also may not.

Part of what ignited this issue was a requirement from a “safety analysis spreadsheet”, that asked the following question regarding stress analysis:

Do the mechanical allowables used for stress analysis have a “sound basis in-fact”, when traced thru all proposed processing steps from raw material, cutting/machining, heat treatment, finishing, etc. IE: does the raw material selection and subsequent processing lead to a rational and obtainable set of mechanical properties?

In my particular case, I gave an example where an existing airframe was analyzed with [equivalent to] “A” basis allowables in a critical beam-cap. In this case the “A” allowables are conservative. However, for the “next generation heavy-weight version” of the same acft, instead of re-designing the beam cap, it was determined that ONLY guaranteed “B” basis allowables would meet the minimum requirement. This was confirmed by structural testing. Subsequent lab-testing of EACH candidate piece of material, provided a rational basis for segregating the premium materials [“B” basis] from ordinary “spec minimum” [“A” basis] raw materials. The beam part number on the “heavy-weight” version was subsequently changed to accept ONLY the premium material… whereas the “standard-weight” version could use either the spec-standard or the premium beam-cap P/N.

At this point in our conversation the stress guy chimed-in that he ONLY used “B” basis in his analysis. I then lead him thru the MMPDS allowables and then pointed spec-minimum allowables (for several typical aluminum alloys)… confirming that spec minimums matched the “A” basis allowables, ONLY. The rest of the afternoon was spent surveying various engineers.

One response that was received after my original posting was different still [stress engineering manager]:

For most secondary and multi-load-path structure the issue is “moot”, since most critical structure has to be analyzed into buckling and permanent deformation to be “proven good for ultimate loading”… and… almost all material out-performs the spec minimums in this analysis [all-be-it twisted-up like a pretzel].


Regards, Wil Taylor

 

[737eng]

In regards to your example with the A-frame. I have seen this type of call-out for material in the commerical world (i.e. in the parts list, the mat'l called out specific mat'l properties that the part mat'l had to meet) and have had some bad experience with this type of call-out from the repair side. In one instance, I was reviewing a repair that was developed by a pretty thorough engineer and noticed that the substantiating data did not take into account the higher allowables req'd in the parts list, it just took the standard allowables (A or B basis can't remember at this time) straight from the MMPDS for the standard mat'l call-out, after I reran the numbers, I could not show the repair good. Luckily, the repair was still in process and I was able to coordinate a valid repair and get it installed with minimal set back. The engineer looked up the mat'l type in the aircraft SRM, which just listed extrusion number and mat'l type, it did not list the special mat'l properties req'd. Another instance was similar, the parts list called out the particular allowables that the mat'l had to meet, and it was nearly impossible to find this mat'l. I eventually had to coordinate with Alcoa and have them heat treat this extrusion for us, big money and huge time lost.
One other example was for fuselage skins, the dwg parts list stated QQ-A-250/5 clad sht fuselage skin quality, after fabricating the replacement skin, we questioned whether or not the sheet used met fuselage skin quality and had to research the sheet all the way back through our purchasing dept, the company we purchased the sheet from an eventually to Alcoa and what batch it was run through with in order to find out what quality it met.

Kind of off the subject, but I just wanted to stress that 1) during original design, consider that this part may have to be replaced or repaired in the future and any special requirement may be hard or impossible to meet by the operator.
2) if your the one designing the repair or coordinating obtaining a new part, ensure you check and ensure that the mat'l procurred meets all the req'd specs.

 

[planedr]

My experience has been to always use the A basis.

I like the mixed approach used by 737Eng. But depends on the buyer as far as DER goes, my former DER would not accept B basis. Transport Canada gets fussy when they find out you use that value. Not "conservative" enough.

The skin quality sheet is another issue I have run across from my friends at RSE. I could never prove the sheet met it, skin quality, nor was it defined. The DER eventually bought the repair the way it was.

By the way 99% of my work has been with RSE for the last 6 years.

 

[737eng]

I took a class offered by Boeing a few years back and remembered that they addressed the A and B-basis. I went back and looked through my notes from the Boeing Structural Repair for Airline Engineers class and they taught that when accomplishing a repair that if your joint utilizes four or more fastener rows to use B-basis, if it utilizes three or less fastener rows to use A-basis. I have written in the margins of my notes that this is the approach that Boeing Stress Engineers and Boeing DER's utilize when evaluating or developing a repair.