Limit vs ultimate load

Discussion in 'Aircraft Design / Aerodynamics / New Technology' started by Emiliano, Nov 21, 2019.

Help Support HomeBuiltAirplanes Forum by donating:

  1. Nov 21, 2019 #1

    Emiliano

    Emiliano

    Emiliano

    New Member

    Joined:
    Apr 29, 2017
    Messages:
    2
    Likes Received:
    2
    Location:
    Italy
    Hi everyone, I've got a question that has been bugging me for a while.

    A typical UL limit load is +4 and -2 Gs. These loads are normally placarded and/or reported on the POH as loads that the pilot should care never to exceed, thus they do not represent (thankfully) the actual structural limits of the aircraft.
    And here comes my doubt: ultimate loads are usually reported as +6 -4 Gs: since our aircraft aren't made of glass or any other brittle material, I suppose all stressed components will undergo a certain amount of plastic deformation before literally splitting apart.
    So when they report 4Gs as operational load 6Gs as "ultimate" positive load, I can interpret it 2 ways, please tell me which one is correct:

    Scenario 1:
    Load test @4Gs: no yield, no failure.
    Load test @ 5.5: Gs: no yield, no failure
    Load test @ 6Gs: some yield seen (e.g. bent bolts, elongated holes) but the structure can still "stay together" perhaps another 0.5 Gs or more

    OR

    Scenario 2:
    Load test @4Gs: no yield, no failure
    Load test @5.5Gs: some yielded components are acceptable (e.g. bent bolts, elongated holes) but the structure must still "stay together"
    Load test @6Gs: the structure must stay together for three seconds before "folding up"

    I had always been convinced that Scenario 1 was the correct one until I read BCAR-S for which I copied and pasted the passage below:

    S 305 Strength and deformation

    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. This applies in particular to the control system.

    b) The structure must be able to support ultimate loads without failure for at least
    three seconds. However, when proof of strength is shown by dynamic tests
    simulating actual load conditions, the three second limit does not apply.



    "The structure must be able to support limit loads without [detrimental] permanent
    deformation"
    . :wonder: I'd assume all permanent deformation is somehow detrimental!

    "At any load up to limit loads, the deformation may not interfere with
    safe operation". :wonder:
    I hope they mean "elastic deformation" by this, because if it's plastic it would mean that I could see yield even below the 4G limit?? How "safe" would it be to fly around with bent bolts and elongated holes? :(
     
    proppastie likes this.
  2. Nov 21, 2019 #2

    Chris Young

    Chris Young

    Chris Young

    Well-Known Member

    Joined:
    Feb 19, 2013
    Messages:
    90
    Likes Received:
    34
    Location:
    France
    Scenario 2 is the normal understanding in the industry. I would even say : 4.1G -> some yield is acceptable. Some designers take large margins but not all do, so unless you checked the stress report yourself, don't assume they did.

    No plastification (yield) is allowed before LL (let's say 4G). The non-detrimental permanent deformation are usually understood to include bolt slippage in clearance-fit, etc... not material yield.

    The deformations that must not interfere with safe operations are indeed elastic deformation, since plastic ones are not allowed. (For example flight controls must not get jammed or resticted by the wing deformation at less than LL).

    Wood and carbon fibre are brittle material (in tension especially) and will not yield much before failure, therefore they are usually sized by ultimate load criteria.
     
  3. Nov 21, 2019 #3

    BBerson

    BBerson

    BBerson

    Well-Known Member HBA Supporter

    Joined:
    Dec 16, 2007
    Messages:
    12,200
    Likes Received:
    2,401
    Location:
    Port Townsend WA
    Typical 2024-t3 yield is about 40ksi. And the ultimate yield is about 60 ksi. Which is 1.5 times yield, hence the 1.5 safety factor rule may be based on this fact.
    So that works out nicely for 2024-t3. Other materials would be different.
    The metal work hardens up to ultimate then total collapse after ultimate may begin.
    Of course, the structure can't be allowed to buckle significantly, up to ultimate.
     
  4. Nov 21, 2019 #4

    wsimpso1

    wsimpso1

    wsimpso1

    Super Moderator Staff Member

    Joined:
    Oct 19, 2003
    Messages:
    6,007
    Likes Received:
    3,261
    Location:
    Saline Michigan
    All good stuff above.

    Elaborating a little:
    Up to and at LL, no yield or fiber failure and everything works fine;
    Above LL and below or at UL, yielding is allowed, but nothing can break or collapse;

    Metals commonly used in airplanes will carry loads higher than yield, will plastically deform considerably and absorb a lot of energy before breaking, so this works pretty well. Lots of good history with metal using these rules.

    Go to wood or composites and things change. Wood and composites do have some continued load carrying capacity above first fiber failure, but not usually anything like 50%. On top of that, composites use a FOS of 2.0, not 1.5. So to prevent breaking stuff or collapse at FOS*LL, you end up designed to at least a little above UL and then know that everything is quite safe down at LL.

    The intent of placing composites at an FOS of 2.0 instead of 1.5 is that many composite systems lose some strength over time due to water infiltration. The amount of loss was determined to be solidly less than 1/3 so increasing 1.5 by one third to 2.0 was the decision. And we have a bunch of history now with pretty good results using that rule. Use it and smile.

    Billski
     
    proppastie likes this.
  5. Nov 22, 2019 #5

    PiperCruisin

    PiperCruisin

    PiperCruisin

    Well-Known Member

    Joined:
    Jan 18, 2017
    Messages:
    136
    Likes Received:
    65
    Location:
    Idaho
    I've spent a lot of my career using FEA. It is a neat tool, but can muddy the waters.

    When we talk about yield and ultimate criteria, it is usually validate with a test. How do you know it yielded? Usually one checks for global permanent displacement below some value. FEA may estimate small, local deformation you would probably not detect in test. Material values used are also typically minimums versus what was actually tested. Always make my brain hurt when I start thinking about the variation fuzziness of it all.
     
    wanttobuild and BoKu like this.
  6. Nov 22, 2019 #6

    Dana

    Dana

    Dana

    Super Moderator Staff Member

    Joined:
    Apr 4, 2007
    Messages:
    8,687
    Likes Received:
    3,064
    Location:
    CT, USA
    That's why we use a 150% safety factor.
     
  7. Nov 22, 2019 #7

    BBerson

    BBerson

    BBerson

    Well-Known Member HBA Supporter

    Joined:
    Dec 16, 2007
    Messages:
    12,200
    Likes Received:
    2,401
    Location:
    Port Townsend WA
    No
    Not easy. You can check for a same spot return after load test, but usually the support has some change.
    In the end, many airplanes are designed with an extra margin on the spar caps for fatigue life instead of ultimate load, I think. Paul Schweizer explains this in his book.
     
    wsimpso1 likes this.
  8. Nov 23, 2019 #8

    Mad MAC

    Mad MAC

    Mad MAC

    Well-Known Member

    Joined:
    Dec 9, 2004
    Messages:
    452
    Likes Received:
    194
    Location:
    Hamilton New Zealand
    In addition to the above:
    the 3 seconds originates from the testing of timber spars (they tend to creep quite a bit) but I know people who run structural test programs on metallic airframes and they are strict about only applying the load for 3 seconds only.
    Not all load cases are subject to limit conditions, spinning being one of them, for normal cat, spin recovery is required but it didn't say the airframe had to still be straight.

    In service, the FOS can not be considered to still be as high as the design value.

    Over sized spar caps tend to be an alternative to improved detail design, failure due to fatigue tends to related to poor load transfer into the spar, these points typical be root fittings, landing gear mounts, etc

    Looking at CAR 3 for wooden structures it just calls out 1.5 and design allowables from ANC-18.
     
  9. Nov 23, 2019 #9

    proppastie

    proppastie

    proppastie

    Well-Known Member

    Joined:
    Feb 19, 2012
    Messages:
    3,885
    Likes Received:
    1,133
    Location:
    NJ
    You hope just wrinkles in thin aluminum but my failures under static load test was sudden collapse of structure.
     
    wsimpso1 likes this.
  10. Nov 23, 2019 #10

    wsimpso1

    wsimpso1

    wsimpso1

    Super Moderator Staff Member

    Joined:
    Oct 19, 2003
    Messages:
    6,007
    Likes Received:
    3,261
    Location:
    Saline Michigan
    Yield - Tough to firmly identify:

    First start with definition. Most metals do not have a yield point clearly defined by any marked behavior changes. So yield ends up being defined by some fixed deviation from elastic behavior, like 0.2% plastic strain in a tensile test. If you have strain gauges or deflection gauges and you are hydraulically loading, you can identify what load reached you criteria after you got there...

    In the airframe you can best establish that yeild has been reached by having dimensional reference points on things like the fuselage that you can watch the wingtips from. Remove the load and the tips have moved, you have yielded it someplace...

    In a thoroughly analyzed airframe, you might have identified some regions clearly as closest to yield and can strain gage these places. When you unload the airframe, if a strain gage zero creeps, the structure is yielding there.

    Pops, clicks, snaps and any vibrations in the structure is a clear indication to unload the structure immediately and go looking for what is acting up so you can know where first failures are occurring. Sometimes it is just bolted joints settling, but if riveted or bonded joints have shifted or material has hit first fiber failures below criteria, well, you know where you have issues.

    Now this whole thing of spar yielding... I have written about this quite a bit on composite design, but it applies to metal and wood too. The shear web close to the caps carries not only shear but it also extends or compresses with caps. This combined loading drives some beef up above the simple sizing that some folks apply, and frequently the light way to strength is by beefing up the caps more than the web. You can end up with bigger FOS in the caps this way to get a structure with adequate FOS...
     
    Last edited: Nov 24, 2019
  11. Nov 24, 2019 #11

    BJC

    BJC

    BJC

    Well-Known Member HBA Supporter

    Joined:
    Oct 7, 2013
    Messages:
    9,739
    Likes Received:
    6,526
    Location:
    97FL, Florida, USA
    Way back when, I was relatively new to competition aerobatics. At a contest, I said to one of the top national competition pilot, “When I have worked on the airplane, or not flown for a couple of weeks, then go practice a sequence, it sounds like the airplane is coming apart. After a few flights, I don’t notice it”. He grinned, and replied, “Yes, it is the same for me”.

    Airplanes flown to the limits make lots of noise, especially in the engine compartment, where lots of pinging and popping originates. Seeing the flying wires vibrating in a blur under significant negative g will get your attention. It is good to learn what is normal, and what isn't.

    The movie “Cloud Dancer” has some audio that includes popping and pinging from the engine compartment. It is realistic.


    BJC
     
    Last edited: Nov 24, 2019
    pictsidhe, Pops and wsimpso1 like this.
  12. Nov 24, 2019 #12

    Pops

    Pops

    Pops

    Well-Known Member

    Joined:
    Jan 1, 2013
    Messages:
    7,228
    Likes Received:
    6,117
    Location:
    USA.
    I have experience that in getting in mountain turbulence a couple of times. Getting the wind knock out of you and hurting your back and trying to hold a pillow between my head and door frame with one hand and fly with the other and keeping the dirty side down was the most hoped for . Lots of popping and cracking going on. One time my wife was with me and after that flight we both was sore for a few days. We both said that it felt like we went several rounds in the ring with Hulk Hogan.
     
    BJC and wsimpso1 like this.
  13. Nov 24, 2019 #13

    wsimpso1

    wsimpso1

    wsimpso1

    Super Moderator Staff Member

    Joined:
    Oct 19, 2003
    Messages:
    6,007
    Likes Received:
    3,261
    Location:
    Saline Michigan
    Propastie knows and has the bent metal to prove it.

    We must put structural collapse beyond the Limit Load times the FOS, even when the stresses look modest as can be. When we have issues with either small scale wrinkling or large scale buckling, the basic calculated stresses in the part can be quite low, well below yield. The most common failures are column loading (Shigley, et al) and crippling (Bruhn). When all else fails, we can go full house with Elastic Stability (Timoshenko and others).

    Billski
     
    Pops likes this.
  14. Nov 24, 2019 #14

    proppastie

    proppastie

    proppastie

    Well-Known Member

    Joined:
    Feb 19, 2012
    Messages:
    3,885
    Likes Received:
    1,133
    Location:
    NJ
    What I think about as I get close to wing load test...at limit load I'm afraid to touch it...how do I know if the full span alerion is jamming ? Support it just at that deflection and see if I can move them ? Will limit loads be momentary and jamming if it was to happen is not significant?
     
  15. Nov 25, 2019 #15

    BBerson

    BBerson

    BBerson

    Well-Known Member HBA Supporter

    Joined:
    Dec 16, 2007
    Messages:
    12,200
    Likes Received:
    2,401
    Location:
    Port Townsend WA
    Shouldn't be afraid at limit load. If you don't want to test to ultimate then you should probably set your design limit load lower until you do test to ultimate.
    The full span aileron is likely a problem. The Schweizer 2-33 has the aileron split in two pieces so it can fold at the joint gap under wing bending. It might even jam at 1g.
     
  16. Nov 25, 2019 #16

    wsimpso1

    wsimpso1

    wsimpso1

    Super Moderator Staff Member

    Joined:
    Oct 19, 2003
    Messages:
    6,007
    Likes Received:
    3,261
    Location:
    Saline Michigan
    You have to avoid impact loads while the wing is loaded, but it ought to be fine at Limit Load. Yeah, you do not want to put any part of your body where the load will drive parts if the wing gives up. All of those riveted joints will creak a bit, but they will do that in flight too.

    I would set up the wing with a mockup of the aileron control on the grounded side of the rig. For each test point, including 1 g, you set up the load, load the wing, measure deflection, wiggle the stick for confirmation, then unload the wing and check that it comes back to the unloaded position. Keep a fish scale handy so that if you get aileron binding or springing to one side or the other, you can measure it at both ends of travel and at the middle. Keep lights and mirrors handy too so you can look at the aileron linkage for source of binding. Then you have something to base a decision on if it binds or springs. Remember that in flight both wings will have the same forces, and that the springiness from one will be in opposition to the other, so your control forces for springiness will be the difference between the two, but binding will be twice the binding of a single.

    The issues of wing bending are not just springing and maybe binding the hinges, but also clearance to the linkage as the wing bends or even latching. If you must, you can break the linkage into parts with idlers, or break the aileron into pieces with a link between the pieces.

    But during test, you should prove to yourself that these long flexible wings can be controlled when under load.

    Billski
     
    Last edited: Nov 25, 2019
  17. Nov 25, 2019 #17

    proppastie

    proppastie

    proppastie

    Well-Known Member

    Joined:
    Feb 19, 2012
    Messages:
    3,885
    Likes Received:
    1,133
    Location:
    NJ
    I plan on the aileron attached at load test...first to test limit aileron load on the wing structure, then load the wing......could wiggle the aileron as I increment load the wing. I do anticipate problems at full limit deflection but again wounder if a momentary limit load in a gust is something to worry about aileron movement.....Of course if I take that attitude at what load should I be concerned.....limit is 5.5G obviously 1 g is a no show and a 60 degree bank is 2 g so one would not want problems then but ....?????
     
  18. Nov 25, 2019 #18

    BBerson

    BBerson

    BBerson

    Well-Known Member HBA Supporter

    Joined:
    Dec 16, 2007
    Messages:
    12,200
    Likes Received:
    2,401
    Location:
    Port Townsend WA
    When you do a 4g load or whatever, then block under the tip, so then wiggling the aileron won't cause a collapse.
    2g might be enough, if the rudder can roll it out of a bank. But if it jams, how would you get it out of a 3G bank at 70°?
     
    proppastie likes this.
  19. Nov 25, 2019 #19

    proppastie

    proppastie

    proppastie

    Well-Known Member

    Joined:
    Feb 19, 2012
    Messages:
    3,885
    Likes Received:
    1,133
    Location:
    NJ
    Well I think I see a path for the test....lets see if/where the problems occur.
     
  20. Nov 25, 2019 #20

    Marc Zeitlin

    Marc Zeitlin

    Marc Zeitlin

    Well-Known Member

    Joined:
    Dec 11, 2015
    Messages:
    509
    Likes Received:
    711
    Location:
    Tehachapi, CA
    Push to unload the wings to <1G, roll level, pull out. Then go fix the problem with the ailerons before flying again.
     
    wsimpso1 and BJC like this.

Share This Page



arrow_white