Aeronautic Freek said:
weight start moving outwards when RPM is increased but accelration is inward?
Before this question can even be answered, you need to get clear about two distinctions that you appear to be failing to make. First, you need to distinguish inertial from non-inertial frames; second, you need to distinguish coordinate acceleration from proper acceleration.
The second distinction is easier to see conceptually. Coordinate acceleration is simply the second time derivative of position with respect to whatever frame (inertial or non-inertial) you are using. Thus, coordinate acceleration is frame-dependent. Proper acceleration is acceleration that is actually felt as "weight" (or measured by an accelerometer). Thus, proper acceleration is frame-independent.
Now for the first distinction. In an inertial frame, analysis is easier because coordinate acceleration can only be caused by proper acceleration; i.e., an object will only have coordinate acceleration if it is being acted on by a force that it actually feels as weight--proper acceleration. So the two always go together in an inertial frame.
In the case of the so-called centrifugal clutch, in an inertial frame, when the engine RPM increases, the weight's tangential speed increases, which means it has coordinate/proper acceleration in the tangential direction (from the engine itself, transmitted by the clutch assembly that is constraining the weight to stay in its channel). This increase in tangential speed causes the radius of the circular path the weight follows to increase; but there is no outward
acceleration in the radial direction in an inertial frame, because the change in the radius of the weight's circular path is entirely accounted for by its increase in
tangential speed and hence its
tangential acceleration. The only
radial acceleration is from the force applied to the weight by the spring, which is inward.
You appear to be analyzing this scenario in a rotating, non-inertial frame, and thinking of the weight's motion as purely radial; however, that actually cannot be the case for a rotating frame with fixed angular velocity, because the RPM, and hence the angular velocity of the weight relative to an inertial frame, is not constant. So, for example, if we use a rotating frame with angular velocity equal to the starting (slower) RPM, the weight starts out at rest in this frame; but when the RPM speeds up, the weight moves
tangentially, not just radially, because the weight's angular velocity, relative to an inertial frame, is now faster than the angular velocity of the rotating frame, relative to an inertial frame. So this analysis ends up looking much like the analysis done above for the inertial frame: the weight's tangential acceleration accounts for the increase in its radial coordinate in this non-rotating frame.
If, instead, we use a rotating frame with angular velocity equal to the
ending (faster) RPM, the weight does not start out at rest in this frame; it is moving tangentially, in a retrograde direction (i.e., opposite to the direction of rotation of the frame relative to a non-inertial frame), and the tangential acceleration induced by the RPM change of the engine now causes that retrograde tangential speed to decrease, which again accounts for the increase in the weight's radial coordinate.
However, using non-rotating frames does require observing the distinction between coordinate acceleration and proper acceleration, since they are no longer the same. Even with constant RPM, the weight experiences an inward proper acceleration, due to the spring; and the spring is the
only thing that is providing any force that is felt as weight/proper acceleration in the radial direction. And that is the case regardless of which frame we choose to do the analysis in. So in
all frames, it is true that the only
proper acceleration on the weight is inward. This is why
@Dale answered "yes" to the question you posed in what I quoted above: he was taking "acceleration" to mean "proper acceleration", since that's the only kind of acceleration that is not frame-dependent and therefore is telling us about the actual physics of the situation, instead of about our choice of coordinates. (And it also tells us that the clutch works just fine no matter what frame we use, as
@Dale said.)
You appear to not be using any of the above frames, but to be thinking of the scenario in terms of a rotating frame with time-varying angular velocity, so that the weight only moves radially in this frame. In such a frame, yes, the weight experiences an outward
coordinate acceleration, and this can be thought of as due to the fact that, while the RPM is changing, the inward force of the spring and the outward centrifugal force do not exactly balance. However, it is still true that
only the inward force of the spring is felt as weight/proper acceleration, so the
proper acceleration is still always inward (thought its magnitude, of course, changes as the spring stretches).
In short, if you are going to say that there is an outward acceleration on the weight, you need to do two things that you have not been doing up to now in this discussion: (1) specify the particular non-inertial frame you are using, and (2) specify that you are talking about coordinate acceleration, not proper acceleration.