• Welcome to engineeringclicks.com
  • Has anyone been involved with HALT testing?

    Discussion in 'The main mechanical design forum' started by GarethW, Jul 5, 2011.

    1. GarethW

      GarethW Chief Clicker Staff Member

      Jul 2009
      Likes Received:
      I've been doing HALT (Highly Accelerated Life Testing) this week. Basically it involves thrashing the test specimen to death in an environmental chamber through temperature cycling and vibration.

      Have you been involved with HALT before? What are your experiences? Let's compare notes!
    3. PWASS

      PWASS Well-Known Member

      Sep 2009
      Likes Received:

      I've been involved in designing and performing various performance life tests for hydraulic valves.

      Also flame retardency, noise generation, IP rating and others.
    4. cwarner7_11

      cwarner7_11 Well-Known Member

      Oct 2009
      Likes Received:
      Many years ago, I did a lot of "Accelerated Life Testing" involving a variety of vibration testing, temperature cycling, humidity, etc. The biggest problem was always finding a proper "transfer function" to relate the test parameters to the actual environment the object would be subjected to. For instance, ferrous materials have a pretty well-defined fatigue life, whereas something like aluminum is extremely difficult to predict, because one can have catastrophic failure of aluminum without obvious warning signs. Another great bugaboo was crack growth- back in the day when I was involved, virtually every model for crack growth deviated several orders of magnitude between prediction and reality (I would hope this situation has improved considerably since I last worked in this area). I suspect the problem is that crack growth depends very heavily on crystal structure at the microscopic level, and it is very difficult to generate a predictive model for the macroscopic environment due to the random distribution of the crystal structure. Another issue that I ran in to is that a device might evidence an infinite life below (or above) a particular temperature, but there is a "step" in the failure rate once a given temperature is exceeded. Thus, just testing at elevated temperature can give very misleading results, unless there is adequate understanding of the exact failure mechanism.

      In short, if one does not have a good model of the anticipated failure mechanism, no amount of testing is going to provide reliable predictive data.

    Share This Page