What exactly is centrifugal force

[stevendaryl]

No, inertial forces are NOT simpler. They may seem simpler to people who prefer to memorize formulas instead of understanding them.

Can someone write down what are the equations for inertial forces, and what are the rules for using them? It seems to me that it amounts to this:

1. Write down the equations of motion using inertial Cartesian coordinates.
2. Transform to curvilinear, noninertial coordinates.
3. Note that there are extra terms in the equations of motion that were not present in the inertial Cartesian case.
4. Call these extra terms "inertial forces".

Surely, the last step isn't doing anything for you. Calling them "forces" doesn't help anything. They are different from other forces you're likely to encounter, because they don't depend on the substance an object is made of, and they don't have an equal and opposite reactive force. Calling them forces is a confusion--it's not a simplification. There is nothing that becomes simpler because of that choice of names.

The real confusion that is at the heart of discussions of "inertial forces" is the assumption that, if \stackrel{\rightarrow}{U} is a vector (say, a velocity vector) with components U^i, then \frac{\stackrel{\rightarrow}{dU}}{dt} must be a vector with components \frac{dU^i}{dt}. That's just bad mathematics. It's just not true. It's true for inertial Cartesian coordinates, but not for other coordinates. It's not a "simplification" to assume something that is provably false.

 

[A.T.]

No, inertial forces are NOT simpler...

Then I'm sure you will soon have convinced anyone to stop using them. Let us know when all the books have been revised.

That's just bad mathematics. It's just not true...

It leads to the same quantitative predictions, which is all that matters in physics. The rest is philosophy.

 

[rcgldr]

Change this 2 body system to one where there are no attractive forces, and the two objects are connected by a string and rotate in a circular path about a common center of mass. Both objects exert a reactive centrifugal force on the ends of the string (assuming that the common center of mass is not located within one of the objects, in which case only one end of the string experiences a reactive centrifugal force).

And how would the objects produce a centrifugal acceleration of the ends of the string?

They wouldn't. The net force on each object is inwards. The objects (and the ends of the string) are accelerated "inwards" by the tension in the string. The outwards force exerted by the objects onto the ends of the string is a reaction (to acceleration) force, not a net force, equal in magnitude and opposing the tension at the ends of the string.

 

[sophiecentaur]

Then I'm sure you will soon have convinced anyone to stop using them. Let us know when all the books have been revised.

It leads to the same quantitative predictions, which is all that matters in physics. The rest is philosophy.

This thread reads like something out of Gulliver's Travels, actually. Anyone would think that there is some actual 'reality' in it all. People don't acknowledge that Science is the pragmatic business of predicting things - all the rest is faith.

 

[Dale]

Just throwing in my 2 cents worth.

That is why I prefer the terms "inertial forces" and "interaction forces".

I agree with A.T. here. Fictitious forces do most things that you expect real forces to do, including do work in the non-inertial frame in which they exist.

The true reaction to a centripetal force is another centripetal force.

This can be true in certain circumstances, but it is not generally true.

So some people would take this to mean "Newton's laws only apply in an inertial frame". I don't like that conclusion. If you view them as vector equations, then they apply in all circumstances, not just inertial frames.

Interesting idea. I was aware of this in terms of gravity in GR, but hadn't thought clearly about the advantage for Newtonian physics also. As we discussed in that other thread, I am not convinced about this because of the double-degeneracy of the metric in Newtonian physics. But I haven't looked at it closely (for that same reason).

There is nothing about the "centrifugal reaction force" that causes anything to flee from the centre. Nothing.

This is simply wrong. If you look at A.T.'s little astronaut cartoon, suppose that the astronaut is standing on a section of the floor supported by bolts which can be suddenly cut. If they are suddenly cut then the reactive centrifugal force will accelerate the section of the floor away from the center. It is only the presence of the bolt forces which prevents the floor from fleeing the center under the influence of the centrifugal reaction force.

 

[rcgldr]

Suppose we have two spherical moons orbiting a spherical planet and both moons are directly opposite each other (ie. a line through the moons' centres passes through the planet's centre) on identical orbits. Would you say that that the reaction forces of each moon on the planet are centrifugal?

In this situation, there is no reaction force, because there is nothing to exert a reaction force onto. The only forces are gravitational. The only "outwards" force would be exerted onto the surface of the planet, but that force is due to gravity, not a reaction force. In this situation, the reaction to gravitational force is a change in the path of the moons as they orbit.

If you look at A.T.'s little astronaut cartoon, suppose that the astronaut is standing on a section of the floor supported by bolts which can be suddenly cut. If they are suddenly cut then the reactive centrifugal force will accelerate the section of the floor away from the center. It is only the presence of the bolt forces which prevents the floor from fleeing the center under the influence of the centrifugal reaction force.

A reactive force is a response to acceleration of an object wrt inertial frame. In an inertial frame, once the floor is cut, the astronaut and the floor cease to accelerate, so there is no reactive centrifugal force. In a rotating frame, the reactive centrifugal force also vanishes (the astronaut ceases to exert a force onto the floor), and the fictitious centrifugal force now changes into a combination of fictitious centrifugal and coriolis forces that correspond to an object moving at constant velocity wrt inertial frame, as observed from a rotating frame.

 

[stevendaryl]

Then I'm sure you will soon have convinced anyone to stop using them. Let us know when all the books have been revised.

There are lots of bad ideas that are taught to beginning students of physics that have to be "untaught" to advanced students. Inertial forces is one of them. "Relativistic mass" is another.

It leads to the same quantitative predictions, which is all that matters in physics. The rest is philosophy.

It doesn't lead to ANYTHING. Calling something a "force" when it's not is just bad terminology. It's not an alternative approach to doing physics.

 

[stevendaryl]

This thread reads like something out of Gulliver's Travels, actually. Anyone would think that there is some actual 'reality' in it all. People don't acknowledge that Science is the pragmatic business of predicting things - all the rest is faith.

The claim that "Science is the pragmatic business of predicting things - all the rest is faith" is itself a philosophical position, and is therefore, not science.

 

[A.T.]

It doesn't lead to ANYTHING.

It leads to the same quantitative predictions, which is all that matters in physics. The rest is philosophy.

 

[stevendaryl]

It leads to the same quantitative predictions, which is all that matters in physics. The rest is philosophy.

Calling something a "force" doesn't lead to ANYTHING. If you think otherwise, give an example of how something follows from the fact that you call certain terms "forces".

 

[stevendaryl]

It leads to the same quantitative predictions, which is all that matters in physics. The rest is philosophy.

The claim that nothing matters other than quantitative predictions is itself a philosophical claim. It's funny that the people who bring up "that's just philosophy" as an argument are the ones who actually end up making the strongest philosophical claims.

 

[A.T.]

The claim that nothing matters other than quantitative predictions...

...in physics.

It's funny that the people who bring up "that's just philosophy" as an argument...

The argument is "The rest is philosophy, so there is no point arguing about it on a physics forum".

 

[stevendaryl]

The argument is "The rest is philosophy, so there is no point arguing about it on a physics forum".

But that's basically this entire thread. The physics part is nothing more than:

"If one uses noninertial, curvilinear coordinates, then additional terms appear in the equations of motion."

One sentence. Everything else is an argument for a particular way of looking at those additional terms.

 

[A.T.]

give an example of how something follows from the fact that you call certain terms "forces".

All forces are just "certain terms". So I don't see how this is an argument against inertial forces specifically.

 

[stevendaryl]

It leads to the same quantitative predictions, which is all that matters in physics. The rest is philosophy.

I also want to point out that connection coefficients (the preferred, in my opinion, way to deal with noninertial, curvilinear coordinates) are essential to understanding General Relativity.

 

[stevendaryl]

All forces are just "certain terms". So I don't see how this is an argument against inertial forces specifically.

No, they're not. Real forces have certain properties that the fake forces don't:
(1) They are vector quantities, meaning that they exist in EVERY coordinate system. The components change when you change coordinate systems, but a vector is a geometric quantity that is independent of coordinates.
(2) Real forces have corresponding reaction forces, leading to conservation of momentum.

In contrast, "inertial forces" are artifacts of a particular choice of coordinates. They don't have corresponding reaction forces. They can be made to disappear by choosing the appropriate coordinate system.

 

[A.T.]

In a rotating frame, the reactive centrifugal force also vanishes (the astronaut ceases to exert a force onto the floor),

You can easily find examples where reactive centrifugal force pushes things outwards in the rotating frame. Two blocks on a turntable. An outer light one with high friction. An inner massive one on rollers. The inner block applies a centrifugal interaction force to the outer block, which pushes the outer block away from the center in the rotating frame.

But all of that is not relevant to the "centrifugal"-label for the reason I state in post #64.

 

[A.T.]

Real forces have certain properties...

Give an example of how something follows from the fact that you call them "forces".

 

[sophiecentaur]

The claim that "Science is the pragmatic business of predicting things - all the rest is faith" is itself a philosophical position, and is therefore, not science.

Haha
I suppose I have to take your point because my suggestion is not falsifiable. But I think that the inverse, - i.e. that Science definitely can establish 'real truth'- probably is falsifiable. So far, we have found this as our experience has been that Science, and its models, continuously changes to fit new evidence.
It has to be true that Science endeavors to avoid saying what things 'really are' because there are so many examples of two or more, equally valid 'realities'. (Note, I write "Science" and not 'Scientists' - who are human and fallible and seldom view things without the distraction of some sort of faith).

 

[Andrew Mason]

This is simply wrong. If you look at A.T.'s little astronaut cartoon, suppose that the astronaut is standing on a section of the floor supported by bolts which can be suddenly cut. If they are suddenly cut then the reactive centrifugal force will accelerate the section of the floor away from the center. It is only the presence of the bolt forces which prevents the floor from fleeing the center under the influence of the centrifugal reaction force.

I have to strongly disagree. It is not reactive centrifugal force that will cause the section to move farther away from the centre. It is the fictitious centrifugal force that would cause that (ie. it is inertia - the absence of centripetal force). The reactive centrifugal force disappears immediately as soon as the bolts are cut. This is exactly why the term "reactive centrifugal force" should not be used. It gets confused with the fictitious centrifugal force.

AM

 

[sophiecentaur]

I have to strongly disagree. It is not reactive centrifugal force that will cause the section to move farther away from the centre. It is the fictitious centrifugal force that would cause that (ie. it is inertia - the absence of centripetal force). The reactive centrifugal force disappears immediately as soon as the bolts are cut. This is exactly why the term "reactive centrifugal force" should not be used. It gets confused with the fictitious centrifugal force.

AM

But the other astronaut (sitting in the frame of the wheel) will see the departing astronaut accelerating, initially (during the first 90 degrees of motion, at least) and due to the geometry of the situation. Would he not conclude that there is a force still operating? This perceived force will also be making the departed astronaut perform a spiral outward path - so it would (might) not just be a centrifugal force that he would need in order to explain the guy's path.

 

[stevendaryl]

Give an example of how something follows from the fact that you call them "forces".

Of course, it doesn't matter what you call them, but the point is that Newton's laws relate motion of one object to vector quantities produced by other objects:

m \frac{\stackrel{\rightarrow}{dU}}{dt} = \stackrel{\rightarrow}{F}

The left-hand side is a fact about the motion of the object, and the right-hand side is about the external situation affecting that motion. In terms of coordinates:

(\frac{\stackrel{\rightarrow}{dU}}{dt})^i = \frac{dU^i}{dt} + sum over j, k of \Gamma^i_{jk} U^j U^k

where \Gamma^i_{jk} are the so-called "connection coefficients", which are due to using nonconstant basis vectors. So the full equations of motion, in terms of components, are:

m(\frac{dU^i}{dt} + sum over j, k of \Gamma^i_{jk} U^j U^k) = F^i

What the idea of "fictitious forces" amounts to is moving the extra terms from the left side (where they describe motion) to the right side (where they are treated as forces):

m \frac{dU^i}{dt} = F^i + F_{inertial}^i

where
F_{inertial}^i = - m sum over j, k of \Gamma^i_{jk} U^j U^k

What difference does it make whether you group it on the left side, or the right side? Well, for one thing, when it comes to figuring out the reaction forces (Newton's third law), only the F^i term is relevant. There are no reaction forces to F_{inertial}^i. For another, since real forces are vectors, the components transform in a standard way under a coordinate change: If you change coordinates from x^i to y^b, then

F^b = sum over i of \dfrac{\partial y^b}{\partial x^i} F^i

"Inertial forces" DON'T transform that way.

So sure, you can group whatever terms together you want, and call them whatever you want to call them, but when it comes to reasoning about the physics, you have to separate out the "real" forces from the "inertial" forces. You're basically doing extra steps that have to be undone later.

 

[stevendaryl]

Haha
I suppose I have to take your point because my suggestion is not falsifiable. But I think that the inverse, - i.e. that Science definitely can establish 'real truth'- probably is falsifiable.

I tend not to talk about truth, one way or the other. I prefer to discuss the ideas without worrying too much about whether they are truth, or something in our heads, or what.

 

[dextercioby]

"Inertial forces" as you (stevendaryl in post #78) mention are a step forward from Newton's postulates (which always have the <with respect to an inertial reference frame> text in them) to Einstein's General Relativity, parallel in a way to the step in which you replace Newton's postulates to Einstein's ones in Special Relativity.

 

[sophiecentaur]

I tend not to talk about truth, one way or the other. I prefer to discuss the ideas without worrying too much about whether they are truth, or something in our heads, or what.

Then I guess you are not far from being a Real Scientist.

 

[Dale]

The claim that nothing matters other than quantitative predictions is itself a philosophical claim. It's funny that the people who bring up "that's just philosophy" as an argument are the ones who actually end up making the strongest philosophical claims.

I don't think it is a philosophical claim, I think it is a semantic claim. I.e. "physics" is defined as X, Y is not X, therefore Y is not "physics".

 

[stevendaryl]

I don't think it is a philosophical claim, I think it is a semantic claim. I.e. "physics" is defined as X, Y is not X, therefore Y is not "physics".

Okay, but who gets to define what the word means? It seems to me that the meaning is provided by watching what physicists actually do, rather than how they would answer question "What is physics?"

 

[Dale]

A reactive force is a response to acceleration of an object wrt inertial frame.

No, the reactive force is an equal and opposite reaction to the other force in a 3rd law pair. This is Newton's 3rd law, which seems to confuse people in rotating frames for some reason. It has nothing to do with acceleration of a specific object since the acceleration of the interacting objects can be different and the acceleration depends on the net force rather than the individual forces.

In an inertial frame, once the floor is cut, the astronaut and the floor cease to accelerate, so there is no reactive centrifugal force.

Sorry, I guess I didn't my proposed scenario clearly. I specified that the bolts were "suddenly" cut for a very important reason. As the astronaut is standing on the floor the floor is under stress with centripetal forces from the bolts and a centrifugal reaction force from the astronaut. The centripetal force is greater than the centrifugal reaction force so there is a net acceleration towards the center.

When the bolts are suddenly cut the stress is relieved from the outside of the section of floor, but the inner part of the floor (where the astronaut is standing) is still under stress. This sets up a shear wave where the floor material transitions from stress to stress-free. During the time between when the bolts are suddenly cut and when that shear wave reaches the feet of the astronaut the centrifugal reaction force still exists, the feet and floor are still in contact, and the floor is accelerating in a direction away from the center. It may help to think of the floor as being made of a stretchy rubber material.

The centrifugal force is every bit as "centrifugal" as the centripetal force is "centripetal". The centrifugal force points away from the center, the centripetal points towards the center. If either is unbalanced then it will result in acceleration in the corresponding direction. If there are other forces involved then the actual acceleration depends on the net force, per Newton's 2nd law.

In a rotating frame

I intended to discuss the reactive centrifugal force only from the perspective of the inertial frame in order to avoid any possible mix-up with the fictitious (inertial) centrifugal force in the rotating frame.

 

[Dale]

The reactive centrifugal force disappears immediately as soon as the bolts are cut.

Hi Andrew Mason, obviously my description was poor since rcgldr had exactly the same response. Please see my response to him in the post above.

 

[Dale]

Okay, but who gets to define what the word means?

Physicists, in particular, the subset of physicists who write physics textbooks. (Maybe we can include Webster and other dictionary writers too)

It seems to me that the meaning is provided by watching what physicists actually do, rather than how they would answer question "What is physics?"

That would be true if physicists did nothing besides physics. However, since physicists do other things besides physics, the definition must come from the answer to the question you posed. That way you can distinguishing between when they are doing physics and when they are doing things that are not physics.

 

[stevendaryl]

That would be true if physicists did nothing besides physics.

Granted. Here's an analogy: sports. Athletes are people who play sports. Of course, an athlete does things besides play sports, but I don't think that an athlete is any better at defining what a "sport" is than anyone else. They can describe what they do when they play sports. I don't think that physicists have any more insight into what "physics" is than an athlete does about what "sports" are.

Anyway, when Einstein, or Newton, or Schrodinger, or just about any other physicist was engaged in doing physics, it certainly wasn't coming up with formulas that make predictions. They were engaged in the struggle to understand the world. That activity is a big part, I would say the center, of what I consider to be physics.

 

[A.T.]

Of course, it doesn't matter what you call them...

So it doesn't matter if you call certain terms "forces" or not. After all it doesn't change the quantitative result of the calculations. That is my point.

 

[stevendaryl]

So it doesn't matter if you call certain terms "forces" or not. After all it doesn't change the quantitative result of the calculations. That is my point.

My point was that it doesn't make any difference what you call things, but that for reasoning, it gets in the way to lump things together that are different sorts of objects.

 

[A.T.]

This is exactly why the term "reactive centrifugal force" should not be used. It gets confused with the fictitious centrifugal force.

I don't see how you can confuse the two. One is an interaction force that exists in every frame, the other is an inertial force that exist only in rotating frames. The diferences are listed in the table here:
http://en.wikipedia.org/wiki/Reactive_centrifugal_force#Relation_to_inertial_centrifugal_force

Calling it "centripetal" as you suggest, despite the fact that it points away from the center, that would be confusing.

 

[sophiecentaur]

Anyway, when Einstein, or Newton, or Schrodinger, or just about any other physicist was engaged in doing physics, it certainly wasn't coming up with formulas that make predictions. They were engaged in the struggle to understand the world. That activity is a big part, I would say the center, of what I consider to be physics.

That assumes everyone has your view of things. I very much doubt that they were as naive as to think they were actually near a conclusion. You cannot have any knowledge of their motives but you must know that anyone who breaks ground in any of this (since the concept of God given laws has ceased to be taken for granted, at least) can only hope to improve on existing scientific models. Models are not 'truth'; they are statements that can be shown to predict the outcomes of certain experiments.
Your three example Scientists were as fallible and human as the next man in many respects and may well have believed at times that the truth is there but they would have been only too aware that it was their models that were the test of their achievements

 

[Dale]

I don't think that an athlete is any better at defining what a "sport" is than anyone else

I don't know the athletic literature very well, but whoever would write a mainstream textbook on sport would be the one who gives the authoritative definition of the word "sport". (or Webster et al.)

what I consider to be physics.

The problem is that if you define "physics" and I define "physics" and we don't agree to use some common definition that we both consider authoritative then we cannot easily communicate since we are using the same symbol with different meanings. That is why we rely on authoratitive sources for definitions of terms, not just each individual's whim. An individual who refuses to use standard definitions and insists on using their own causes all sorts of communication problems.

In any case, we are straying from my main point, which was that a "that's just philosophy" statement is a semantic statement, not a philosophical statement. I wasn't attesting to the accuracy of A.T.'s statement, and I won't debate the merits of different definitions. Semantic arguments are boring because they are always arguments from authority and someone can always refuse to recognize your chosen authority and substitute their own preferred authority (usually themselves).

 

[A.T.]

for reasoning, it gets in the way

Whatever that means...

 

[stevendaryl]

Whatever that means...

I gave plenty of examples of what I meant. If you didn't understand what I meant after that, you could ask follow-up questions.

 

[stevendaryl]

The problem is that if you define "physics" and I define "physics" and we don't agree to use some common definition that we both consider authoritative then we cannot easily communicate since we are using the same symbol with different meanings.

I don't think that's true. The fact that you and I might disagree about specific cases whether something is or is not "physics" or a "sport" or "music" does not get in the way of communication if there is substantial overlap.

 

[stevendaryl]

That assumes everyone has your view of things.

No, it doesn't. If people have different points of view, that's fine--there's room for lots of different kinds of physics. The sort of physics that is done by cosmologists, or loop quantum gravity people, or those working in the foundations of quantum mechanics is very different from the kind of physics that is done in solid state physics or biophysics. There is room for all.

 

[stevendaryl]

I very much doubt that they were as naive as to think they were actually near a conclusion. You cannot have any knowledge of their motives but you must know that anyone who breaks ground in any of this (since the concept of God given laws has ceased to be taken for granted, at least) can only hope to improve on existing scientific models. Models are not 'truth'; they are statements that can be shown to predict the outcomes of certain experiments.

I'm not sure what the relevance of this is to what I've said. Claiming that the goal of physics is understanding doesn't imply how close our current understanding is to the truth.

 

[A.T.]

I gave plenty of examples of what I meant. If you didn't understand what I meant after that, you could ask follow-up questions.

I understand that it is not relevant for the predictions.

Anyway, when Einstein, or Newton, or Schrodinger, or just about any other physicist was engaged in doing physics, it certainly wasn't coming up with formulas that make predictions.

Yes it was.

They were engaged in the struggle to understand the world.

Define "understand the world".

 

[Dale]

I don't think that's true. The fact that you and I might disagree about specific cases whether something is or is not "physics" or a "sport" or "music" does not get in the way of communication if there is substantial overlap.

The current discussion between you and AT seems to contradict this claim of yours. There is surely substantial overlap, and yet the remaining differences in the definitions are getting in the way of communication.

 

[Studiot]

How does any of this help or encourage the OP?

 

[Dale]

I think that there is no such thing as centrifugal force .

Am I right ? is this force fictitious ?

A centrifugal force is a fictitious force. That doesn't mean that there is no such thing as a centrifugal force.

 

[Andrew Mason]

But the other astronaut (sitting in the frame of the wheel) will see the departing astronaut accelerating, initially (during the first 90 degrees of motion, at least) and due to the geometry of the situation.

He would think the departing astronaut was accelerating only if he forgot that he was in a non-inertial (rotating) reference frame. To an inertial observer, the departing astronaut is simply continuing the motion he had when the bolts were cut.

Would he not conclude that there is a force still operating? This perceived force will also be making the departed astronaut perform a spiral outward path - so it would (might) not just be a centrifugal force that he would need in order to explain the guy's path.

I think the departing astronaut would prescribe a cycloidal outward spiral in the non-inertial reference frame of the astronaut on the rotating space station.

AM

 

[stevendaryl]

The current discussion between you and AT seems to contradict this claim of yours. There is surely substantial overlap, and yet the remaining differences in the definitions are getting in the way of communication.

I don't think that's a correct diagnosis.

 

[stevendaryl]

I understand that it is not relevant for the predictions.

Indirectly, it is. Getting straight the difference between connection coefficients and forces is an important step in understanding physics in curved spacetime. So if you want to go on to advanced topics, then understanding this is important.

Define "understand the world".

Come on. You know what the word "understand" means.

The idea that physics is exclusively about making quantitative predictions is just wrong.

We can go through many examples. Before Einstein developed Special Relativity, the equations of Special Relativity were already developed. That's why they're called the "Lorentz transformations" rather than the "Einstein transformations". Einstein's new theory didn't change the equations, it provided a new way of understanding those equations--a new way of deriving them. In the long run, this was a tremendous advance, making most of the physics since then possible. But the motivation wasn't new predictions, it was to understand things that people already knew about, but didn't understand.

Similarly for quantum mechanics. The Balmer series already gave a good quantitative prediction for the energy levels of hydrogen. It was just a guess. The steps that Bohr took, which opened up further developments by Schrodinger and Heisenberg, was an attempt to derive those energy levels from some kind of first principles. Of course, quantum mechanics ended up having enormous consequences and great predictive value. But the initial steps were an attempt to understand already existing information.

When Feynman developed his path integral formulation of quantum mechanics, at first it was simply a reworking of the Schrodinger--it was just a different way of understanding how to derive quantum amplitudes. It turns out that the ideas developed by Feynman in working on his path integral formulation had applicability beyond quantum mechanics, and could be applied in quantum electrodynamics and elsewhere.

The idea that there is nothing to physics other than quantitative predictions is a philosophical position, and in my opinion, it's very BAD philosophy. The attempt to understand data and formalisms has always been the most direct route to coming up with new theories that do make quantitative predictions. So even if, at the end, all you care about is quantitative predictions, trying to understand the mathematics, the data, and the models is a much more effective way to get to that point.

 

[Dale]

The idea that there is nothing to physics other than quantitative predictions is a philosophical position, and in my opinion, it's very BAD philosophy.

No, it is a definition of the term "physics", and in your opinion it is a very BAD definition.

 

[stevendaryl]

Some other examples: General Relativity. The motivation was to make a theory of gravity that was compatible with relativity. There was also a philosophical goal, which was to generalize the principle of relativity so that all coordinate systems were treated equivalently, not just inertial coordinate systems. The goal wasn't to make new predictions about bending of starlight, or whatever. The goal was reconciling two theories: Special Relativity and gravity. It certainly turned out that the project produced new predictions, and if hadn't produced new predictions, it would been considered a failure, or at best an interesting exercise. However, the goal of making predictions really didn't drive the development of the theory at all. Einstein was trying to understand the nature of gravity in a way that made sense in light of relativity.

I just think that the idea that there is nothing to physics other than quantitative predictions is just a severely claustrophic notion of what science is about. It is true that at the end of the day, your ideas have to have empirical consequences, but a lot of the development of a new theory is about clarifying concepts.