Fitting factor and casting factor 1

[gantovnik]

Follow to the discussion from Link

Fitting factor FF = 1.15

In [14 CFR § 25.625 - Fitting factors] we can read.
"No fitting factor need be used with respect to any bearing surface for which a larger special factor is used."

Assume, I have casting factor CF=1.25, because the part is made by casting. Should I remove fitting factor? Or CF=1.25 is belong to material properties/cert, and not that "larger special factor" mentioned by FAA.

Thank you!

Vladimir Gantovnik, PhD

http://gantovnik.com

 

[Sparweb]

You don't have to combine the FF with the CF when analyzing the properties and strength of a casting. Let the CF apply to material strength properties, and the FF apply to fasteners.

Note, some analysts will do their analysis by inserting the FF or CF directly into the material properties, or at other times, include these factors in the applied loads. I don't know why they do that, and it seems confusing and wasteful to me. I find it much more efficient to use the loads and stresses as they are determined by analysis, and use allowable strength values as they are published or determined by test. Only at the last step when determining a Margin of Safety, to include the appropriate Factor(s).

You can find a very detailed discussion about special factors, including these and more, in the last chapter of E.F. Bruhn's text book Analysis and Design of Flight Vehicle Structures (if you can find one).

 

[rb1957]

there is a separate FAR for casting factors, 621 maybe. There are a range of casting factors, depending on the process and the criticality of the part.

you don't Need to combine FF and CF, but an extra 15% conservatism ... would that break the bank ?

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[gantovnik]

Thank you, guys. Of course, I know the current practice in our industry, and we use only one (largest) factor, CF.

But... let's see the origin/background of these factors FF and CF:

Origin of FF: "Many uncertainties exist concerning the stress distribution in fittings. Manufacturing tolerances are such that bolts never fit the holes perfectly, and small variations in dimensions may affect the stress distribution. An additional margin of safety of 15 percent for military airplanes and 20 percent for civil airplanes is used in the fittings design". So manufacturing tolerances, variations in dimensions... (Ref. Peery, D.J., Aircraft Structures, McGraw-Hill, 1959)

Origin of CF: "The application of safety factors to castings is based on the fact that the casting process can be inconsistent. Casting is a method of forming an object by pouring molten metal into a mold, allowing the material to solidify inside it, and removing it when solidification is complete. Castings are subject to variability in mechanical properties due to this casting process, which can result in imperfections, such as voids, within the cast part." (Ref. FAA)

Now, assume that my supplier is terrible. I received a part in my fitting with the following: 1) errors in manufacturing tolerances and variations in dimensions, 2) variability in mechanical properties, imperfections, and voids in casting. We have two different types of errors: apples and pears. The question now is, why would you use a statement from the FAA, "No fitting factor needs to be used with respect to any bearing surface for which a larger special factor is used," to eliminate FF and use CF only?

See what I mean?

Vladimir Gantovnik, PhD

http://gantovnik.com

 

[rb1957]

There's nothing "Wrong" with adding a factor to cover minor manufacturing defects (in hindsight), or of scrapping the part because it fails to meet the drawing requirements, or of scrapping 'cause if the re-analyze with the final part you get a -ve MS.

There's nothing "Wrong" with adding an extra conservative factor (in the early analysis) to increase coverage of the kind of thing you mention.

The Casting Factor takes into account the manufacturing process and the part's criticality. You have a pretty low factor (1.25) compared with others I've seen (2), so I think you have a good controlled process and a non-critical part ?

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[SWComposites]

Now, assume that my supplier is terrible. I received a part in my fitting with the following: 1) errors in manufacturing tolerances and variations in dimensions, 2) variability in mechanical properties, imperfections, and voids in casting. > This is not an analysis issue. a) reject the parts, b) get a new supplier.

 

[gantovnik]

No, no! This is just an assumption about the bad supplier and two errors. I don't know if that part has geometrical defects (covered by FF) and property defects (voids) covered by CF. But these two errors could happen at the same time! And instead of applying CF=1.25 on material properties and the separate application of FF=1.15 for geometry error, we use only CF=1.25 because "No fitting factor needs to be used with respect to any bearing surface for which a larger special factor is used." CF is not part of the set of those "larger special factors" mentioned by the FAA. I feel it belongs to material properties only and should be applied separately. Thank you very much!

Vladimir Gantovnik, PhD

http://gantovnik.com

 

[Ng2020]

Careful-
A fitting factor reflects uncertainty in the internal stress state (i.e. complex/uncertain internal loads) and may be removed if proof of strength for the detail is shown by test.

You can find its origins in ANC-1, with some more flowery wording, from memory.

Such uncertainty may arise if you have a complex internal load state, usually at points of load introduction eg in a lug. Sufficiently far enough away from points of load introduction, this uncertainty diminishes ( St Venants principle ).

Fitting Factor does not act as a stop-gap for out of tolerance parts, poor manufacturing quality.

In this aspect, the FAA deemed it is too conservative to apply both a conservative 'load uncertainty factor' and a conservative 'material /process variability factor' simultaneously. (Presumably they deem the likelihood of these worst case scenarios coinciding to be extremely improbable, or something along those lines)

 

[WKTaylor]

FYI related...

FAA AC 25.621 Casting Factors

Regards, Wil Taylor
o Trust - But Verify!
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", HBA forum]
o Only fools and charlatans know everything and understand everything." -Anton Chekhov

 

[gantovnik]

I will add this description of the fitting factor from JSC 65828. Contrary to the description given by the FAA (14 CFR § 27.625 - Fitting factors), it contains a very good rationale for the fitting factors (and it is more modern than the description by Peery, D.J., Aircraft Structures, McGraw-Hill, 1959):

Vladimir Gantovnik, PhD

http://gantovnik.com

 

[rb1957]

yes, the FF is for concentrated loadpaths (ie, fittings) where ther can be small changes in loads due to small changes in the distribution of a limited number of loadpaths.

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[Ng2020]

Historically...
See CAR3.303 "variability factors" ( = 'special factors' in today's parlance), which states the purpose of these factors is to ensure that the probability of a part being understrength is 'extremely remote'.
I.e It's a probability argument, intended to maintain a desired level of safety.
Hence the combination of multiple factors (casting x fitting) is unnecessarily conservative.

 

[WKTaylor]

FYI ONLY...

14 CFR Ch. I
25.337 Limit maneuvering load factors.
25.527 Hull and main float load factors
25.619 Special factors.
25.621 Casting factors.
25.623 Bearing factors.
25.625 Fitting factors.
25.1531 Maneuvering flight load factors.

Regards, Wil Taylor
o Trust - But Verify!
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", HBA forum]
o Only fools and charlatans know everything and understand everything." -Anton Chekhov

 

[SWComposites]

Well, frankly, if a casting is a flight critical part, then to me it must be tested at full scale, and therefore the fitting factor is not required anyway. And if the part or supplier is so terrible, then that part should not be a flight critical part in the first place.

 

[WKTaylor]

For Giggles...

WL-TR-91-4111 Durability and Damage Tolerance of Aluminum Castings

Army RE-71-2 FATIGUE PROPERTIES OF INVESTMENT CASTINGS

FAA ANM-112N-13-05 Casting Factors [Replaced by AC25.621-1?]

AGARD AG-299 Handbook on Advanced Casting

AGARD-CP-325 Advanced Casting Technology

AGARD-R-763 Castings Airworthiness

Regards, Wil Taylor
o Trust - But Verify!
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", HBA forum]
o Only fools and charlatans know everything and understand everything." -Anton Chekhov

 

[WKTaylor]

SWC... There are 'premium quality' strength/toughness/soundness aluminum alloy castings available... using special alloys, casting practices and HT... followed by very high quality NDI and continuous random sample testing.

Likewise for titanium and certain alloy steels.

30 Years ago I was at an F-16 Conference and LMTAS was chewing-over weapons pylon re-designs. Off handedly I mentioned that the F-15 weapons pylons were premium quality Titanium castings... made the F-16 guys suddenly think about doing that design possibility. I think their canopy frames were already premium cast aluminum alloys.

Heck NOW days additive manufacturing has advanced so quickly... I suspect there are huge advantages to consider.
/NOTE1/ There are dozens of MIL, ASTM, company, etc... process, and material specs on the topic of aerospace premium quality additive manufactured airframe parts.

Regards, Wil Taylor
o Trust - But Verify!
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", HBA forum]
o Only fools and charlatans know everything and understand everything." -Anton Chekhov

 

[rb1957]

thee days why cast ? additive machining should be able to accomplish what shape you need.

For me, if cast, use the highest quality you can afford, and apply a conservative factor (something like 2). If it's a detail part then no weight impact; if it is a much repeated part then weight could be an issue.

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[mfgenggear]

Casting are can be a pita. And a lot can change from one pour to the other. Strict quality is required. Dimensionally, quality of pour, alloy.
There can be issues very quickly. Some one has to take ownership and continue to monitor and inspect. On top of the quality sample requirements.

 

[rb1957]

you said it ... "castings are a PITA".

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.