Approach for determining jerk in a mechanical system

[MayZ91]

Hello all,
I require your expertise on building an approach or a framework to identify and measure the jerk or vibration in a mechanical system with a slider.
What I have is a slider (S) with a vertical link (L1) attached to it and another horizontal link (L2) attached at the tip of the link (L1).
I would like to understand what approach to take in identifying the jerk experienced by the above system when slider S starts to move at an acceleration (a).
I only need to understand the approach to take, it could be a mathematical modeling and simulation or testing. But a step by step procedure alone is sufficient, I can execute the project from that point on wards.

I understand the description is very vague, but this is all the information I have and I am expected to gather the inputs that are required to solve. Any suggestions or critics are most welcomed, since there is very less information that I have provided.

 

[Nidum]

(1) The question is so vague that no one on PF can sensibly answer it .

(2) Theoretical or practical methods could be equally appropriate . Have you made any attempt to find out what methods are available and to decide which would be the best to use ?

(3) Jerk and vibration are different things .

No one on PF wants to be unhelpful though - try to establish a better definition of the problem , do some research , propose one or more solutions and then come back and tell us what you have done ..

If you have questions about specific detail aspects of this problem you are welcome to post these any time

 

[BvU]

Hi May,
Perhaps it's a good idea to make a sketch of what you yourself have in mind for this setup, approximately ? I am guessing this is in preparation for an assignment or lab experiment ?

 

[Dr.D]

I think the process can be outlined, even without the diagram.

1. Draw the vector loop diagram, showing the connected vectors that define the configuration;
2. Write the equations corresponding to the diagram;
3. Differentiate these equations three times with respect to time.
The first differentiation gives the velocity relations; the second differentiation gives the acceleration relations, and the third differentiation gives the jerk relations. Solve these systems of equations (closed form or numerical solution) and that should answer your question.

 

[JBA]

Or, simply use: F=ma. Essentially, the force that it will take to accelerate the part is the "jerk" force reaction that the part will experience.

 

[Dr.D]

Actually, what JBA proposes is not correct. "Jerk" is a technical term and means the third derivative of a motion. Only in very exceptional circumstances could the third derivative be directly proportional to the second derivative, the acceleration = a. While 'jerk" may seem to describe the acceleration in terms of perceptions, that will not do since there is a generally accepted technical definition.

 

[JBA]

I stand corrected now that I have seen the definition of "jerk" as a formal technical/physics terminology. In some 50 years of mechanical/machinery engineering design I have never seen "jerk" applied in its technical/physics sense.

 

[Dr.D]

It is certainly true that jerk is not discussed nearly as often as position, velocity and acceleration. I have encountered it primarily in connection with the design of automotive cams.

 

[FactChecker]

Without see a diagram, I would comment that the jerk is a function of how fast the acceleration builds to a. You have given nothing to indicate that. If you suppose that the acceleration goes from 0 to a instantaneously, then the jerk will be infinite.

 

[JBA]

To see a detailed discussion, curve examples, and application examples of Jerk, including your below concern see the below:

https://en.wikipedia.org/wiki/Jerk_(physics)

If you suppose that the acceleration goes from 0 to a instantaneously, then the jerk will be infinite.

 

[newjerseyrunner]

I'd like to see a diagram of what you are trying to do. If it's a simple mechanical system, can't Autocad do most of the calculations for you? If you're trying to eliminate jerkiness with a redesign and aren't sure how, personally, I'd use a genetic algorithm and simulate a thousand generations or so.