Why Is the Moment of Inertia Not Constant in Motion Control?

[zoltrix]

hello

take a mechanism driven by a DC motor
in automation books you find transfer functions such as :
G(s) = k1/s(s+k2)
normally it s assumed that the moment of inertia referred to motor shaft is a constant
the point is that in general it is not a constant
take for example a slider - crank or a 4 bar mechanisms
so I wonder what is the point of such sophisticated studies ?

 

[Lnewqban]

... so I wonder what is the point of such sophisticated studies ?

What studies are you referring to?

 

[jrmichler]

The better your understanding of the theory, the better your ability to deal with real world problems.

Dealing with load inertia is especially fun when the system has backlash, and the load inertia is zero when in the backlash. That's the worst case of variable load inertia. There is a solution that works, and you need to know enough theory to understand why that solution works in order to properly design the solution.

Another real world problem is torsional rigidity in a servo system. You need to understand the theory well enough to know why a properly designed finite jerk motion profile is a complete solution. And the side effects and limitations of that solution.

But first, you need to understand the theory of a system with constant load inertia, infinite stiffness, and a simple DC motor. Only then will you be ready to learn how to tackle real world challenges.

 

[zoltrix]

What studies are you referring to?

I am referring,in general, to "automation and control engineering" applied to mechanical engineering
the theory assumes linear systems but mechanical systems in general, are not linear
take for example the transfer function , which I quoted in my previous post , for a
DC motor + mechanical load system
it is assumed that the the moment of inertia of the load referred to the shaft of the DC motor is constant
however even for common and simple linkages the equivalent inertia is far away for being constant
The methods of linearization of non linear systems are definitely too much complicated for a real use
so my question is :
is it it worthwhile , for a mechanical egineer , studying "automation an control engineering " ?

 

[zoltrix]

The better your understanding of the theory, the better your ability to deal with real world problems.

Dealing with load inertia is especially fun when the system has backlash, and the load inertia is zero when in the backlash. That's the worst case of variable load inertia. There is a solution that works, and you need to know enough theory to understand why that solution works in order to properly design the solution.

I dont think itis just a matter of backlash
the moment of inertia referred to the shaft of, slider-crank mechanism is not constant because of the geometry of the linkage

 

[Baluncore]

the moment of inertia referred to the shaft of, slider-crank mechanism is not constant because of the geometry of the linkage

Maybe you should balance the mechanism to reduce vibration and hopefully stabilise the MoI to a range that can be controlled.

 

[zoltrix]

put it an other way
take a slider-crank mechanism
you want to move the slider at constant velocity over a certain period of time
consequently you must properly control the torque of the DC motor
my question was :
can "automation and control theory " be of use for pratical "motion control" problems even in case of non linear systems ?

 

[Baluncore]

... you want to move the slider at constant velocity over a certain period of time
consequently you must properly control the torque of the DC motor ...

If the controller does not know where the slider mechanism is now, how can it control anything?

can "automation and control theory " be of use for pratical "motion control" problems even in case of non linear systems ?

Probably yes.

Each post, in this thread, is feedback in a control system. If you use the English language without punctuation, we cannot be sure what you mean, and it would be pointless for you to study control theory because your reasoning is irrational, like an open-loop.

 

[zoltrix]

punctuation ? come on...maybe I could not make myself understood since English is not my native language...

anyway
Of course the controller must know the position of the slider i.e. the time interval/stretch of path at constant velocity
I dont think it is the main issue
the point ,in my opinion ,is the non linearity of the system

can anybody suggest a book / text dealing with motion control of mechanical systems ?

 

[Lnewqban]

...can anybody suggest a book / text dealing with motion control of mechanical systems?

https://www.amazon.com/s?k=motion+c...i=stripbooks&ref=nb_sb_noss&tag=pfamazon01-20