1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Difference between torque and moment

  1. Jun 1, 2009 #1
    Torque is when a couple is formed...and for a moment, a couple needs not be there right?
     
  2. jcsd
  3. Jun 1, 2009 #2

    tiny-tim

    User Avatar
    Science Advisor
    Homework Helper

    Hi dE_logics! :smile:

    "torque" sometimes means a couple, and sometimes not. :rolleyes:

    From the PF Library on https://www.physicsforums.com/library.php?do=view_item&itemid=175"

    Torque is the moment of a force about a point.


    "A torque" is also the name of a pair of equal-and-opposite forces which are not in-line, and so have a purely rotational effect.


    "Torque" vs. "moment":

    The words "torque" and "moment" (of force) mean the same.

    However, "torque" tends to be used when there is an axle or pivot to be turned around, while "moment" tends to be used in essentially non-rotational situations, such as analysis of forces on a beam.
     
    Last edited by a moderator: Apr 24, 2017
  4. Jun 1, 2009 #3
    You topple if you get down from a rapidly moving bus because of the torque exerted by friction alone..
    Moment and Torque are not always the same. Moment is a blanket term. Moment of a force is torque. Moment of momentum is angular momentum. Moment of a vector X, is R x X.
     
    Last edited: Jun 1, 2009
  5. Jun 1, 2009 #4
    Although torque and moment of a force are the same, engineers tend to use moment (I believe an older term) as a way of suggesting motion (a cognitive sticky, if you will).
     
  6. Jun 2, 2009 #5
    You mean in case the moment by each force is not balanced, then it will make a torque.

    So moment is only in statics and it turns to torque in kinetics.

    The 3 answers sound pretty confusing.
     
  7. Jun 2, 2009 #6

    tiny-tim

    User Avatar
    Science Advisor
    Homework Helper

    You seem to be expecting English to have the same rigour as maths. :rolleyes:

    In English, "torque" is a bit vague, and in particular, there is a difference between torque of a force and "a torque".

    Don't worry! :blushing: … the context usually makes it clear. :smile:
     
  8. Jun 2, 2009 #7
  9. Jun 3, 2009 #8
    :uhh: Lets just talk science here........


    No 100% English vocab.

    BTW how do you even relate it to the English torque?...or other meanings?
     
  10. Jun 3, 2009 #9
    Balanced or unbalanced, moment of a force about a point is called torque. All you need is a force and a reference point. A force doesn't cease to exist just because it is neutralized by some other force.
     
  11. Jun 4, 2009 #10
    Yeah, that's why we have rotation.

    So torque and moment are the same thing.

    Or is it that, when there is no equilibrium, then its torque, else moment.
     
  12. Jun 4, 2009 #11
    A torque is a torque is a torque.. :smile:
     
  13. Jun 4, 2009 #12

    D H

    User Avatar
    Staff Emeritus
    Science Advisor

    Moment of force is a synonym for torque. They mean exactly the same thing. Using the term "moment", sans the "of force" qualifier, is physicists being a tad lazy. There are lots of other "moments". In addition to those already mentioned, moment of inertia.
     
  14. Jun 8, 2009 #13
  15. Feb 16, 2011 #14
    Moment is the tendency for a force(s) to create rotation about a point.

    In Physics: Moment is Torque is Moment. Done.

    In Engineering: Moment is Moment. Torque is the Moment of a Couple.

    A Couple is a Moment with ZERO NET FORCE.
    A Moment of a Couple can be moved anywhere on the body and cause the same rotation. It is thus sometimes called a "Free Vector". Though this may seem non-intuitive, it is a real characteristic of a Couple.

    "Equilibrium" is not an intrinsic necessity to qualify a Moment or a Couple. If there is rotational motion, there is definitely no Moment equilibrium (sum of Moments = 0). However, if it's a Statics context, there would definitely be translational equilibrium. If there is a resultant Force or Moment, there is spin, and thus no equilibrium. If you are BALANCING a system, then you will set forces and moments to zero and then FIND equilibrium. I hope that clears it up.
     
  16. Feb 16, 2011 #15
    This sems to be a very old thread to resurrect, although the discussion is perennial.
     
  17. Feb 17, 2011 #16
    True, Studiot. I'm in a Statics class and came across this thread just yesterday (and several other similar). There seems to be no real closure to some of them (uncontested misinformation, missing information, questions unanswered) so I thought I'd at least clear this one up.
     
  18. Feb 17, 2011 #17
    There are two real differences between a moment (or couple) and torque.

    Firstly Torque is not limited to a single revolution.

    Secondly moments (and couples) are planar beasts - they exist in a plane. Torque, on the other hand, is three dimensional and has the ability to transfer moments from one plane to another.
     
    Last edited: Feb 17, 2011
  19. Feb 17, 2011 #18
    Moments/couples certainly exist in 3-D. Their resultant vector projects into the third dimension. This is no different than a Torque.

    Couples transfer moments the same as Torques, being free vectors.

    Is a Moment limited to a single revolution? Please, show me where or how.
     
  20. Feb 17, 2011 #19
    Perhaps you can display a non planar couple?

    If the two forces constituting the couple are not in the same plane how can there be a zero force resultant?

    A single force (line) and a single fulcrum (point in space)can only be planar.
     
  21. Feb 17, 2011 #20
    Yes. But we're not talking about a Force and a point, we're talking about a Force and a Position vector, with origin at a point. If we're not talking about two vectors, then we can't be talking about a Moment or a Couple.
     
  22. Feb 17, 2011 #21
    Every force has a moment about every single point in space.

    You can have a position vector if you like, but it does not add anything except complication.

    To create a moment all you require is a point and one single force.

    Elementary geometry prescribes that these be coplaner, since you can always establish a plane containing a given line and a given point.

    If you introuduce a second non concurrent force to the system you can create a couple.

    If you apply this couple to a material body you can cause rotational motion or bending.

    Bending is not torque.

    If you apply a second couple to the same body you can cause torsion or power transfer. This is the three dimensional effect I was referring to. This is how screwdrivers, socket sets and engine shafts work.
     
  23. Feb 17, 2011 #22
    Every force has a moment about every single point in space.

    No, a force CAN create a moment about any convenient point on a rigid body when the distance between this point and any point of application along the line of force is represented as a position vector. This representation is not just pedantic, it represents the physical connection between an applied force and an axis. If you just have a point in space and a force in space, then you just have a point in space and a force in space. They don't interact.
    This is all described mathematically (and physically) and conforms to vector space rules, including the vector (cross) product. This can be found in any Statics or Physics book. If you can point me to any evidence to the contrary, I'd love to see or be directed to it.

    You can have a position vector if you like, but it does not add anything except complication.

    No, it's not that I can, but I MUST have a position vector. Conversely, YOU can have a point and a vector, but you won't have a torque, moment, or couple. Holding onto the idea that a point and a vector (force) is all that's needed for a moment is what is adding the complication.

    To create a moment all you require is a point and one single force.

    Just stating this again does not make it any more valid. You need two vectors to create a Moment.

    Elementary geometry prescribes that these be coplaner, since you can always establish a plane containing a given line and a given point.

    Elementary geometry prescribes that two parallel forces that don't reside along the same line of action are indeed coplaner, the key term here being Forces, not Moments. A Moment requires a third dimension, for the resultant moment vector. Again, constricting a Moment to 2-D is really just another way of saying the MAGNITUDE of a moment.

    If you introuduce a second non concurrent force to the system you can create a couple.

    No, you MUST introduce a second force to create a couple, but that's not enough, the force also has to have equal magnitude, opposite direction, and not reside along the line of action of the first force. Just having another force be concurrent is not enough, actually.

    If you apply this couple to a material body you can cause rotational motion or bending.

    Absolutely it CAN cause rotation (providing there is not equilibrium). I'm not sure why you're introducing bending moments. This is a discussion about Moments, Torques, and Couples.

    Bending is not torque.

    And Bending doesn't qualify a Moment.

    If you apply a second couple to the same body you can cause torsion or power transfer. This is the three dimensional effect I was referring to. This is how screwdrivers, socket sets and engine shafts work.

    I don't disagree here.

    I can point you to countless references corroborating my understanding of a Moment, including the maths involved. They will all explicitly necessitate a position vector, and not just a point. This vector is most certainly not a "complication", it is a strict necessity. Here is one:

    http://www.engin.brown.edu/courses/en3/notes/Statics/moments/moments.htm

    I'll provide more if you ask.

    Again, if you can point me to any definition of a moment that does not require a position vector, I would love to see it. If you can describe, physically, a Moment involving a point in space and a force, and nothing between them, I'd love to hear it. If you can describe mathematically how to get a perpendicular resultant moment vector, mathematically, from a plane containing a point (scalar) and a vector, then please, I'd love to hear it.
     
    Last edited by a moderator: Apr 25, 2017
  24. Feb 18, 2011 #23
    That was a big post, with many points to discuss.

    Remembering the thread title I will take the last one first.

    If we have a shaft and apply a single couple to it we simply have a rotating shaft. No torque is involved.

    If we apply a second equal counter acting couple coaxially then the shaft will not rotate it will twist.

    This is torque.

    I think you agree with this?

    Alternatively we can apply a lesser couple (brake?) and extract work.

    I am sorry to disillusion you but a bending moment is a true moment.
    If you take the case of the root an encastre cantilever, there are two support reactions.
    One of these is a moment, not a force, nor yet a couple. In other cases a bending moment may be the result of a couple.

    With regard to vectors, vectors are not necessary to consider moments. And moments certainly do not need three dimensions. Take the calculation for the centroid of an L shaped lamina, so beloved of mechanics exam questions. The point in question is in free space, and not part of the body at all. Yet we take moments about it.
    Nor is the third dimension required for a rotational vector, which incidentally is not a true vector at all, since it does not obey the commutative law of addition.

    In my experience and if one has learned one's lessons well at a particular level, one of the most difficult things to do is to unlearn these lessons when proceeding to the next level. I found this particularly so in the case of vectors.

    go well
     
  25. Feb 18, 2011 #24
    That was a big post, with many points to discuss.

    It was big because you are kitchen-sinking.

    Remembering the thread title I will take the last one first.

    If we have a shaft and apply a single couple to it we simply have a rotating shaft. No torque is involved.

    No, we have a shaft and a couple. Rotation is not necessary, we still have a torque and a moment. We have a body, acted upon by forces in the configuration of a Couple. Because it happens to be rotating changes nothing. There is mass, and it's being acted upon by rotational (as opposed to translational) forces. There IS torque/moment. And it DOES include a position vector. These are the kind of posts that add to the confusion. This is why I posted in the first place.

    I think you're probably trying to use specifically an example of a FRICTIONLESS SHAFT. Ignoring that there is no such thing in reality, still, you are rotating a body when you apply a couple. When analyzing the system, the shaft would be included in any free body diagram. If that body has mass, then you have Torque (couple). There is no getting around that. If there is no body (mass) then you have nothing to apply force to. This is a simple concept, and building up complication does not make your argument work any better. In fact, you made my argument for me: a couple applied to a shaft made the shaft rotate. It had to have torque/moment. You would agree now, right?

    If we apply a second equal counter acting couple coaxially then the shaft will not rotate it will twist.

    Sure, it CAN twist. And you don't even need a second couple.

    This is torque.

    Well, yes. There was torque/moment in the first place. If we apply another torque, sure there is still torque.

    I think you agree with this?

    Yes, i agree if you have torque and you apply a torque, there can still be torque. Yes, we agree!

    Alternatively we can apply a lesser couple (brake?) and extract work.

    Sure.

    I am sorry to disillusion you but a bending moment is a true moment.

    Don't be sorry, i'm quite illusioned! You introduced the qualifier "Bending". I don't disagree that this is a moment, but it's a distinction, a certain way to look at a moment. My point was, there is no need to introduce Bending Moment into our discussion, as Moment did just fine, and it has not changed your argument. A torque is still a moment.

    If you take the case of the root an encastre cantilever, there are two support reactions.
    One of these is a moment, not a force, nor yet a couple. In other cases a bending moment may be the result of a couple.

    Show me an example, online.

    With regard to vectors, vectors are not necessary to consider moments.
    Yes, they are. I've given you an example, as evidence. Please, show (not tell) me yours, won't you?

    And moments certainly do not need three dimensions.

    Yes, they do. Moments NEED 3 dimensions. It is in the example i've give you. Show me yours.

    Take the calculation for the centroid of an L shaped lamina, so beloved of mechanics exam questions. The point in question is in free space, and not part of the body at all. Yet we take moments about it.

    Not exactly. The point is the center mass (of a plane), and the moment can be taken using this (or any arbitrary) point. Still, a Position Vector is taken from this point. The real physical interaction is actually happening along the L-shaped body. The force is being transferred along it. You still HAVE to have a vector. You still HAVE to have something for the force to act upon. A dot and a force in space are just a dot and a force. Please point me to an example of a centroid with a force acting upon it resulting in a Moment but NOT containing a POSITION VECTOR.

    Nor is the third dimension required for a rotational vector, which incidentally is not a true vector at all, since it does not obey the commutative law of addition.

    A third dimension is necessary for a Moment. It has to. A moment vector, not so incidentally, IS in fact a vector. I covered this in "pseudo vector" above. The math still works. We just use conventions for the direction of it's projection. Please, point me to an example of a Moment which DOES NOT result in a Moment vector pointing in a dimension not occupied by the plane of the forces in which the Moment was created. Please. I beg you.

    In my experience and if one has learned one's lessons well at a particular level, one of the most difficult things to do is to unlearn these lessons when proceeding to the next level. I found this particularly so in the case of vectors.

    I agree with you here. That is why I posted here originally. It seems particularly important to post accurate and specific, as well as factual, information about math/physics/engineering on of all things a Physics forum. I have, and still do, source them, and know well the frustration of misinformation.
    I've pointed you to a website supporting my claims. Please, show me/us some evidence supporting your claims. I don't think that's too much to ask. If you're right, there should be a multitude of reputable sources out there you could quickly link to.
     
  26. Feb 18, 2011 #25
    This isn't a p__ing contest.

    It is supposed to be an opportunity in learning and understanding.

    So if you really don't understand how an encastre cantilever works, simply ask and I will explain.

    Do you have any idea what a concurrent force is or what the implications of such a force might be, because the above response suggests otherwise?

    For all you know the shaft might be floating freely in space. I did not specify any other forces/agents/moments acting quite deliberately.

    Because I wanted to make the point that if the only such agent you apply to said shaft is a single couple, it must perforce rotate indefinitely. Your jumble of a response has completely turned that round.

    So, no I don't agree, I stand by what I originally said.
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook