iSixSigma

MSA – Measurement Loosening and Tightening Torque

Six Sigma – iSixSigma Forums General Forums Methodology MSA – Measurement Loosening and Tightening Torque

Viewing 11 posts - 1 through 11 (of 11 total)
  • Author
    Posts
  • #54405

    Jozef
    Participant

    I have a question how to measure Loosening and tightening torque – MSA study.
    My problem is:
    Loosening or tightening torque measurement – there is no possible repeability measurement.
    example: If you will broke screw (tightening torque), u cannot re-measure same part again.

    How do you measure or what equipment is used?

    Any idea is welcome.
    Thanks in advance.

    Problem is that if operator unscrew part, loosening torque cannot be measures?

    Which device is the best for have a data from loosening torque and tightenning torque?

    I need it due VDA -MSA analysis.

    Any ideas is welcome.

    0
    #195244

    Chris Seider
    Participant

    If you have a process that is automatically screwing on caps, nuts, etc., you can select 2 consecutive ones or use 2 consecutive ones by a certain “head” that’s applying the torque. The assumption is the device is fairly consistent and if you get repeatable measurements doing the MSA, then you know all is well with your MSA.

    0
    #195245

    MBBinWI
    Participant

    @jovoga – yes, when you torque a bolt, you stretch it, but unless you are over torqueing (going beyond the yield point) you should be OK re-using the bolts.
    If you are concerned, you can use the same lot of bolts and only use them once, but increase the repeats to dampen out any internal variation of each specific bolt. And if it is very critical to have absolutely precise understanding, then you can do a metrology analysis on the bolts post measurement and if any deviate significantly from the others, you can remove that data point as an anomaly.

    0
    #195248

    The solution we found many years ago on CIJP nozzles was to use Dorian Shainin’s method of build and breakdown. The trick is to understand the tightening torque is not the same as the un-tightening torque – I think this is called hysteresis. When I mentioned this a few years ago on isixsigma I was flamed by several anonymous posters .. but I’m hoping someone from the Cambridge inkjet industry will be able to verify my claim :-)

    0
    #195250

    Chris Seider
    Participant

    Did you find pink X’s or red? :)

    0
    #195252

    Chris, there is only one Red x and one Pink – you need to brush up on your Shainin Techniques :-)

    0
    #195253

    Chris Seider
    Participant

    Pale and Pink I equate the same lol…..

    0
    #195258

    MBBinWI
    Participant

    @Andy-U – Andy: Of course tightening and loosening torque are going to be different (tightening involves dynamic resistance and loosening involves static resistance – the difference being keeping something moving vs. getting something to move). What is really important is the static resistance, as once you get a fastener tightened you are more interested in what it will take to get it to loosen up. However, we usually measure the tightening torque, which is dynamic resistance. So, when the clamping of a joint is critical, then it is best practice to add additional clamping force (more bolts) or if that isn’t possible, then to study the torque to tighten vs. the torque to loosen to establish the proper tightening torque spec.
    Hope this adds to the knowledge.

    0
    #195259

    @MBBinWI Is static resistance the same as stiction?

    Just to confirm my understanding, the problem is to set to a particular torque and then the measure the torque setting. Our method measures the torque set by measuring the breakdown torque, which from memory was slightly lower; so the question then became given the offset how can we measure repeatability? Now you may not agree with our approach, but we used Shainin’s isoplot and accepted the bias between time 1 and time 2.

    0
    #195260

    MBBinWI
    Participant

    @Andy-U – yes, that is a new term that has been applied to static friction (resistance).

    Not sure exactly what you are calling breakdown torque (could this be “break-away” torque? If so, then it should be higher than dynamic torque). What happens is that surfaces in contact form bonds at the atomic level, essentially welding themselves together. To break these bonds, it takes more force than when two surfaces are already moving relative to one another (once the nut or screw is turning).

    In the instance of bolted joints, you also have the mechanism of bolt stretch that is also providing a force in the system. As the joint is loosened, the force of the stretch is reduced as well.

    Repeatability of bolted joints is predominantly a function of the materials, the geometry, and the torque. If you use parts from the same lot, the first two are fairly well controlled – then you measure the third. Again, if you need to dampen out the variation of the parts, just up the repetitions.

    0
    #195261

    OK .. we’re on the same wavelength. Your statement “In the instance of bolted joints, you also have the mechanism of bolt stretch that is also providing a force in the system. As the joint is loosened, the force of the stretch is reduced as well.” resonates with my experience; but my memory is a little vague now.

    I’ve used the terminology favoured by Shainin when performing his component search procedure. Before swapping components, he recommends building and breakdown down the assembly five times to check the repeatability of response.

    0
Viewing 11 posts - 1 through 11 (of 11 total)

You must be logged in to reply to this topic.