The Zenair Cricket (CriCri) Flutter debacle

Discussion in 'Aircraft Design / Aerodynamics / New Technology' started by MadRocketScientist, Jul 6, 2019.

Help Support HomeBuiltAirplanes Forum by donating:

  1. Jul 6, 2019 #1

    MadRocketScientist

    MadRocketScientist

    MadRocketScientist

    Well-Known Member

    Joined:
    May 17, 2009
    Messages:
    1,405
    Likes Received:
    853
    Location:
    Canterbury, New Zealand, Earth, Milky Way Galaxy.
    Rather than continuing to hijack the CH 640 thread here >>
    https://www.homebuiltairplanes.com/forums/threads/vne-limit-on-ch-640.31867/
    I think its best to continue this part of the discussion in its own thread.

    There was quite a mess in the late 1970's when CriCri kits were being produced and sold in North America by Zenair. There was an agreement with the original designer Mr Colomban, but Zenair made many changes to the design in the process of producing the kits. Long story short there was a fatality due to flutter in the modified aileron controls.

    The CriCri has a mixer for the flapperons and one of the critical parts is a 32mm diameter 2024 aluminium tube that runs under the seat pan. This was changed to a smaller 4130 steel tube, presumably to reduce complexity and speed up the manufacturing time. Turns out to be a terrible idea when someone just uses TLAR instead of actually sitting down and running the numbers.
     
  2. Jul 6, 2019 #2

    MadRocketScientist

    MadRocketScientist

    MadRocketScientist

    Well-Known Member

    Joined:
    May 17, 2009
    Messages:
    1,405
    Likes Received:
    853
    Location:
    Canterbury, New Zealand, Earth, Milky Way Galaxy.
    While I have known about the problems with trying to get CriCri plans in USA and Canada for as long as I have been building my CriCri, the CH640 thread prompted me to look at it a little deeper.

    I don't have exact drawings and sizes for the Zenair modified flapperon tube, but I do have scans of the A4 photocopied sheets that were added to the kit plans. With the relevant "drawing" copied into my CAD program and scaled to size, the modified tube seems to be closest 3/4" diameter. cockpit1.jpg
    Here is the cockpit layout.
    DSC00540 (Medium).JPG
    This is the flapperon tube, I sourced this one from Aircraft Spruce and is 1-1/4" diameter and .035" wall thickness.
    Flapperon link tube.JPG
    This is the Zenair 4130 steel flapperon tube partially wrapped in brown paper
    Super_Cricket_05 (Medium).jpg
    You can see the ends sticking up in these two pictures (red arrows). These pictures are from G-ZOW, the Zenair Cricket that had a pair of Rotax 277's installed. That didn't bode to well for the vibration levels and it didn't seem to do much flying.
    Super_Cricket_06 (Medium).jpg

    One thing that only just struck me is that with the design change of the flapperon tube, the leverage force is doubled, due to the longer 'arms' on the ends of the flapperon link tube. Consequently the steel tube needed to be almost twice as stiff as the aluminium one to resist flutter to the same degree.
     

    Attached Files:

    Last edited: Jul 6, 2019
  3. Jul 6, 2019 #3

    MadRocketScientist

    MadRocketScientist

    MadRocketScientist

    Well-Known Member

    Joined:
    May 17, 2009
    Messages:
    1,405
    Likes Received:
    853
    Location:
    Canterbury, New Zealand, Earth, Milky Way Galaxy.
    And now for the exciting part (I jest;)) the numbers...

    The standard flapperon tube is 32mm diameter with a 1mm wall thickness and the length used is 408mm end to end (Centerline of the rivets on the fittings) The modulus of elasticity of 2024 aluminium is 73GPa. Tracing the terrible plans of the modified flapperon tube in CAD, I came up with a length of 421mm for the 4130 tube and the leverage applied by the flapperons due to the longer arms, is 1.93 times the plans version. Using these numbers in an online calculator I did a comparison comparing the torsional stiffness of the 2024 and the 4130 steel tubes.

    Lets call the 2024 aluminium plans tube as 100% torsional stiffness.

    I decided to check the 2024 Imperial tube that I am using as it is slightly smaller diameter and thickness at 31.75x0.889mm (1-1/4x0.035"). The torsional stiffness of this tube is 87.7% of the plans. The next thicker wall thickness, 0.049", would be 119% so in hindsight maybe the thicker wall tube would have been preferred.

    The 4130 tube however, is dismal.:mad:

    Assuming a wall thickness of 0.035", and allowing for the longer leverage and a worst case modulus of elasticity of 190GPa, the torsional stiffness of the 4130 tube is 23.5% of the plans tubing.

    Increasing the wall thickness to 0.065" only brings the torsional stiffness up to 38.6%:eek::eek: Even a solid 3/4" bar of 4130 would only be 72% of the torsional stiffness of the original tube.

    Just for kicks, the size of the 4130 tube needed would be right between 1-1/8" and 1-1/4" with a 0.035" wall to have the same torsional stiffness as the original flapperon tube.

    Considering the modified tube was under 1/4 of the torsional stiffness of the design, it isn't really any surprise that there was flutter on several aircraft.
     
    Vigilant1 and BJC like this.
  4. Jul 6, 2019 #4

    TFF

    TFF

    TFF

    Well-Known Member

    Joined:
    Apr 28, 2010
    Messages:
    11,112
    Likes Received:
    3,011
    Location:
    Memphis, TN
    And probably why the one hanging in the EAA gift shop is hanging in the EAA gift shop.
     
  5. Jul 6, 2019 #5

    pictsidhe

    pictsidhe

    pictsidhe

    Well-Known Member

    Joined:
    Jul 15, 2014
    Messages:
    5,853
    Likes Received:
    1,484
    Location:
    North Carolina
    I found this article last night while looking into the cricket. It seems very strange that the lawyers did not chase the kit manufacturer.
     
  6. Jul 6, 2019 #6

    Tiger Tim

    Tiger Tim

    Tiger Tim

    Well-Known Member

    Joined:
    Apr 26, 2013
    Messages:
    2,767
    Likes Received:
    1,533
    Location:
    Thunder Bay
    When did the lawsuit happen? I ask because I distinctly remember picking up Zenair brochures at aviation shows in the early nineties that still advertised the Cricket. To be fair, they also advertised the Zipper and I’m not sure if a single kit for that one was ever produced.
     
  7. Jul 6, 2019 #7

    mcrae0104

    mcrae0104

    mcrae0104

    Armchair Mafia Conspirator HBA Supporter

    Joined:
    Oct 27, 2009
    Messages:
    2,858
    Likes Received:
    1,882
    Location:
    BDU, BJC
    If you are using the elastic modulus instead of the shear modulus to calculate the torsional stiffness of a tube, then the results are wrong.
     
  8. Jul 6, 2019 #8

    Vigilant1

    Vigilant1

    Vigilant1

    Well-Known Member

    Joined:
    Jan 24, 2011
    Messages:
    3,550
    Likes Received:
    1,527
    Location:
    US
    Yes. Zenair is a small outfit, and may have few assets to satisfy a judgement after all the liquidation costs. Many kit manufactures carry no insurance, so there's no pot of gold there. It is possible that Colomban ( like many successful professionals) might have been a juicier target, and might have even had some type of liability insurance /umbrella coverage that sweetened the deal. If his personal assets weren't effectively walled off from claims against his business activities (which is not as easy as many people think, esp in international claims), the plaintiff's lawyer was just going for the deepest pockets.
     
    Last edited: Jul 6, 2019
  9. Jul 6, 2019 #9

    wsimpso1

    wsimpso1

    wsimpso1

    Well-Known Member

    Joined:
    Oct 19, 2003
    Messages:
    5,621
    Likes Received:
    2,866
    Location:
    Saline Michigan
    Not so you would notice - true that GJ/l is the right figure of merit for a tube loaded in torsion, but G is a function of E and Poisson's ratio, which does not change much from metal to metal, and J is just twice I in cylindrical tubes. The big difference in stiffness due to the slenderness of the steel part will still dominate and in about the same ratio as EI/l.

    Billski
     
  10. Jul 6, 2019 #10

    pictsidhe

    pictsidhe

    pictsidhe

    Well-Known Member

    Joined:
    Jul 15, 2014
    Messages:
    5,853
    Likes Received:
    1,484
    Location:
    North Carolina
    Indeed, but the ratio between shear modulii for aluminium and steel is pretty similar to the ratio of Young's moduli. The relative stiffness is going to be about the same.
     
  11. Jul 6, 2019 #11

    pictsidhe

    pictsidhe

    pictsidhe

    Well-Known Member

    Joined:
    Jul 15, 2014
    Messages:
    5,853
    Likes Received:
    1,484
    Location:
    North Carolina
    Don't they usually go for everyone's pockets?
     
  12. Jul 6, 2019 #12

    Vigilant1

    Vigilant1

    Vigilant1

    Well-Known Member

    Joined:
    Jan 24, 2011
    Messages:
    3,550
    Likes Received:
    1,527
    Location:
    US
    Regarding the lawsuit: We are (so far) working with scraps from some highly partisan recollections. I'd say it would be very hard to know if/how damages against Zenair were pursued. Even knowing the actual parties to a lawsuit wouldn't tell us much.
     
    philg likes this.
  13. Jul 6, 2019 #13

    Pops

    Pops

    Pops

    Well-Known Member

    Joined:
    Jan 1, 2013
    Messages:
    6,661
    Likes Received:
    5,539
    Location:
    USA.

    The reason to stay out of an aviation business, even if you win, you lose.
     
  14. Jul 6, 2019 #14

    mcrae0104

    mcrae0104

    mcrae0104

    Armchair Mafia Conspirator HBA Supporter

    Joined:
    Oct 27, 2009
    Messages:
    2,858
    Likes Received:
    1,882
    Location:
    BDU, BJC
    Yes, that's right. I thought the math was further off because the OP factored the lever arm into the stiffness and I missed that. It's not so much that the steel section is one quarter as stiff; it's closer to half as stiff, and subjected to twice the torque (assuming the speculation about the Zenair sizes is correct).

    $$JG_{orig}= {\pi \over 32} (1.260^4 - 1.181^4) (4,060,000) = 227,360$$
    $$JG_{zen}= {\pi \over 32} (0.750^4 - 0.680^4) (11,600,000) = 116,000$$
    $${JG_{zen} \over JG_{orig}} = {116,000 \over 227,360} = 51\%$$
     
    Vigilant1 and MadRocketScientist like this.
  15. Jul 6, 2019 #15

    pictsidhe

    pictsidhe

    pictsidhe

    Well-Known Member

    Joined:
    Jul 15, 2014
    Messages:
    5,853
    Likes Received:
    1,484
    Location:
    North Carolina
    Unless you're a lawyer. They win, even when they lose.
    Perhaps if they didn't get paid when they lost a case, they wouldn't be such *&&%^&^%
     
  16. Jul 6, 2019 #16

    Vigilant1

    Vigilant1

    Vigilant1

    Well-Known Member

    Joined:
    Jan 24, 2011
    Messages:
    3,550
    Likes Received:
    1,527
    Location:
    US
    The majority of the lawyers handiling these types of cases don't get paid if they lose, they work strictly on contingency. But, the payoff is big enough and the manhours they have at risk are small enough, that it is still worth giving it a shot (and the folks they are attacking still have to mount an expensive defense). Some version of "loser pays" (pays court costs, and the legal costs of the other side) can address this, but that reform also needs to be carefully constructed (else valid lawsuits against big companies become impossible--if I sue Boeing/GM, etc for a valid reason, I've got no control over what they spend to defend. If I lose, I get slaughtered by paying for their very expensive lawyers.) But there are surely ways to improve over the lottery system we have now.
     
  17. Jul 7, 2019 #17

    MadRocketScientist

    MadRocketScientist

    MadRocketScientist

    Well-Known Member

    Joined:
    May 17, 2009
    Messages:
    1,405
    Likes Received:
    853
    Location:
    Canterbury, New Zealand, Earth, Milky Way Galaxy.
    So I used the wrong units but got the right answer :D

    The Cricket 'may' have gotten away with the change if the lever arm length had stayed the same, reducing the stiffness to around 1/4 really was a big mistake.
     
    Vigilant1 likes this.
  18. Jul 7, 2019 #18

    MadRocketScientist

    MadRocketScientist

    MadRocketScientist

    Well-Known Member

    Joined:
    May 17, 2009
    Messages:
    1,405
    Likes Received:
    853
    Location:
    Canterbury, New Zealand, Earth, Milky Way Galaxy.
    I have attached a pdf that has timelines of most of the Zenair designs up till 2008.
     

    Attached Files:

    • DH.pdf
      File size:
      151.2 KB
      Views:
      17
    Tiger Tim likes this.
  19. Jul 7, 2019 #19

    Pops

    Pops

    Pops

    Well-Known Member

    Joined:
    Jan 1, 2013
    Messages:
    6,661
    Likes Received:
    5,539
    Location:
    USA.
    I have a Zenith 600 less the canopy and firewall forward. Engine mount for a small Cont. Plans show a 1835 cc VW engine. Maybe the 2180 flywheel drive VW engine that I have been working on.
     

    Attached Files:

    Last edited: Jul 7, 2019
  20. Jul 8, 2019 #20

    Scheny

    Scheny

    Scheny

    Well-Known Member

    Joined:
    Feb 26, 2019
    Messages:
    52
    Likes Received:
    13
    Location:
    Vienna, Austria
    Here in Austria and most of Europe we have this system that the looser pays. For criminal charges you can ask for a free lawyer if you cannot afford one, but in civil court you can loose big time. For civil cases the starting party has to pay up front all fees, so most of the time you wouldn't go to trial unless defending party has lots of money.

    There are few loopholes, like when some activists "demonstrated" against the wrong fashion shop (owner friend of government). The state attorney made the trial that expensive, that although trial won, the activists left with half a million in debts after getting the standard fee for his attorney.
     

Share This Page

arrow_white