Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Failure of a connecting rod

  1. Apr 9, 2005 #1
    Hey Guys,
    Please help me out.
    Q. The connecting rod of a newly developed internal combustion engine for ultra-light vechicle failed catastrophically during testing, damaging the crankshaft and smashing the engine block into two. An emamination of the fracture surfaces revealed that the cause of failure was fatigue and that the crack had initiated from the root ofone of the conrod threads.
    I've begun my own investigation by carrying out some preliminary finite element analysis. By doing so, l was able to find out the distribution of hoop stress around surface of pin hole.

    I need to write a full-detailed report about too many points but am mostly struggling with
    1- cause of cracking,
    2- any recommended modifications.
    3- justification for the recommended modifications.

    Please help me.I am seeking for details on those points.
    Many thanks

  2. jcsd
  3. Apr 9, 2005 #2
    Cracks propogate along planes of max stress and do so from pre-existing flaws. Bolt threads are man-made pre-existing flaws--each of those threads is a failure waiting to happen. That is why connectin rods are not threaded. Rod bolts are splined along their axis, or they have a very round high pitch ridge or they use 'waves' like these: http://www.twoguysgarage.com/img/misc_prods/260/arp_connecting_rod_bolt.jpg [Broken] to secure the bolt into the con rod. The lower rod cap slides over the threaded bolt and a nut is then used to hold everything together. So, if your rod is threaded then you might consider changing that.

    If you decide to keep the threads then you will want to use a rounded thread design to minimize stress concentration. This is a bad thread: http://www.jeilmed.co.kr/img-all/cli-90.files/image005.jpg

    Another option is to user a stronger material (I use Ti rods in my bike which allow me to run less valve-piston clearance).

    Another option is to use heli-coils. Heli-coil taps(if take the time and spend the extra money for them) come in various designes which produce rounded thread cuts instead of the 'V' shaped ones used by most threads. http://www.emhart.com/products/helicoil/techtopic/tt4.htm [Broken] The rounded cuts minimize the stresses and the heli-coils end up distributing the stresses better than a plain bolt to thread interface. This is one of the reasons a lot of race engines are built with heli-coils from the get-go.

    Anyway, hope this helped a little.
    Last edited by a moderator: May 2, 2017
  4. Apr 10, 2005 #3
    More details needed Plz

    Exellent information faust9 but l really need alot more details.
    As you can see from the attached pic that the max hoop stress takes place at 90 degrees from the y axis i.e. where the red indication is. This shows that cracks are most likely to be initiated in this region.
    So what details can you mention regarding all that?
    What modifications are best recommended to do in this region apart from changing the conrod material to a stronger one i.e. ls it possible to play abit with the existing conrod design?

    l'll be very thankful.

    Attached Files:

  5. Apr 10, 2005 #4


    User Avatar
    Science Advisor
    Gold Member

    To pin the cause of cracking a studying the fracture surfaces in detail and finding out the initiation site of the fatigue crack propagation is pretty much a necessity. After id:ing the fundamental reason (material defect, manufacturing fault, overload etc.) the justification for any further action comes close by. Other than that working on the design to reduce stresses and finding a material better able to handle the fatigue seem like good options (doing a fatigue crack initiation and propagation analyses might not be that bad way to go).
  6. Apr 10, 2005 #5


    User Avatar
    Science Advisor

    First off, good post by Faust. I second the use of radiused root threads. I would recommend looking up the specs for the UNJ classification of threads.

    Secondly, from your analysis (which is really small and tough to see the details), this looks surprisingly like a simple plate with a stress concentration in the middle and opposing loads at the top and bottom. Nothing too spectacular about that. It's following basic theory.

    What material is the part made from? Since you mention light weight, I am assuming it is not a form of steel, but I would hope that it is.
  7. Apr 10, 2005 #6
    I'm still not sure if you threaded the con rod. I'll reiterate my first suggestion--loose the threads. Next, from what I see is the rod is going into tension where the rod and cap meet. That can be eliminated by increasing the con rod bolt pre-load(another reason not to thread con rods is you'd have to worry about stripping these when you increased bolt pre-load). Do some FEA with bolt-preload's. You can reduce the dynamic tensile stresses (tensile=catastrophic failure) at the expense of added compressive stresses (a little compressive yield among friends never hurt anyone).

    As Fred said, look into alternate materials. Ti rods will run from $1500 to 4500 for a 4 cylinder but they are strong. The problem with Ti rods is they fail without warning so using them requires more periodic tear down and inspections(I tear my engine down every year--I used to race--and measure the unloaded rod-bolt length this gives a real good indication of the loads the rods themselves were subject to).

    How long was the engine running and at what RPM before failure? If your rod failed within a few minutes of startup for the first time then you're going to have to redesign them. If they failed after x-hours at 15,500RPM then you might be able to add a maintenance inspection to check the rods for signs of impending failure(ie if your engine failed at 40 hours of continuous 10,000 rpm operation then you'd want to inspect at 15 hours or so--crazy numbers I made up). Following this same scenario, reduce the max RPM.

    What were your minimum flaw assumptions, and did you verify the rod(s) met those assumptions prior to use?

    How were the rods made(cast or forged)?

    What is the piston mass(try reducing this to minimize stresses on the rod)?

    What is your rod ratio? A poor rod ratio can--and will--increase the stresses in an engine.

    Did you inspect the cylinder for abrasions? I've seen failed pin retainers cause rods to snap(chunk of retainer wedged between piston and cylinder bore---SNAP slapping rod cut the case in half).

    Well, there are probably 8587346874659823 other things one could look at to solve your problem. From the picture I would venture a guess that you need to improve your rod bolt design and increase your rod bolt preload.

    Well, good luck with your endeavor.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook