Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Controlled Energy Release Mechanism

  1. Jan 7, 2017 #1
    I am working on a mechanism which controls the release of energy while remaining energy efficient. An image of the concept is shown below. (The image is meant simply to get an idea of how it would work- it is by no means perfect).
    A source of energy (from a spring being released, for example) applies a torque on the yellow crankshaft. This drives the green rack up and down via the red connecting rod. This in turn drives the brown gear. The brown gear rotates the blue gear increasing its velocity, and this blue gear is connected to the purple flywheel. This flywheel is forced to oscillate rapidly. However, due to the momentum of the flywheel, it forces the mechanism to operate at a slower speed. This slows the rate at which the energy is released.
    To smooth out how energy is released, a freewheel can be connected to the yellow crankshaft. Another flywheel would be connected to the freewheel. This flywheel should have a smooth motion.
    I would appreciate any thoughts on whether or not a mechanism such as this would work.
     

    Attached Files:

  2. jcsd
  3. Jan 7, 2017 #2
    You ask if it will work ...I'm not sure what it is you are trying to do ... you say ... 'it's a mechanism which controls the release of energy' , from a spring ...I'm not sure it would do that , give an example of where it would be used .... have a look at a wind up watch , that has a mechanism for the controlled release of a spring
     
  4. Jan 8, 2017 #3
    An escapement of a mechanical watch, while extremely accurate, is not efficient. The mechanical efficiency usually isn't greater than 50% according to this article: http://www.europastar.com/watch-kno...anical-unlocking-alternative-escapements.html. I am attempting to create a mechanism which is much less accurate, making it useless for telling time, but has a much greater energy efficiency- far less energy would be converted into sound and heat.
    I would like the mechanism to slow down the rate in which the spring is released. The energy stored in the spring would be transferred to the rotation of a shaft.
    An example which would use such a mechanism would be a toy car which is wound up with a spring. Instead of the spring moving the toy car in a short burst, it would move it slower over a longer period of time.
    I hope this clears up the confusion.
     
    Last edited: Jan 8, 2017
  5. Jan 8, 2017 #4
    Yes ... that makes things clearer ..... But ...
    So we have the green rack going up and down , but this will just make the brown gear go back and forth , not rotate .
     
    Last edited: Jan 8, 2017
  6. Jan 8, 2017 #5

    Bystander

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member
    2016 Award

    ... and?
     
  7. Jan 8, 2017 #6
    It was my intention for the brown gear to rotate back and forth That back and forth rotation will cause the flywheel to rotate back and forth as well. I admit that I should have been more specific by what I meant when I wrote that it rotates.

    That example was just meant to clarify how this mechanism operates.

    The purpose which I designed it for was a bicycle which can store and release energy in the form of a spring. For such a design, there had to be some way of controlling how fast the spring is released. If it was released all at once, similar to the toy car example, there would be a short burst of acceleration and that would be it- it wouldn't last for nearly long enough. Using friction pads to slow it down, or using an escapement, would not be energy efficient. This mechanism should slow the rate at which the spring is released while losing little energy in the process.

    I've been made aware that bicycles similar to these which use batteries are already on the market, and a battery would be able to store more energy than a spring. However, this design becoming commercially successful seems like a goal so far-fetched at the moment that it is definitely not my main objective. My focus is bringing an interesting concept to life.
     
    Last edited: Jan 8, 2017
  8. Jan 8, 2017 #7
    It will work ,its like a clock work.
     
  9. Jan 8, 2017 #8
    If that is true, that is a great relief. The reason for designing this instead of just using an ordinary mechanism for a clock is to increase the efficiency. Do you believe that my speculation of it having a higher mechanical efficiency is correct?

    This is an image of the mechanism with the freewheel and flywheel contraption which I wrote about before included. In this image the yellow crankshaft is the input while the brown flywheel is the output. The crankshaft rotates the blue freewheel. The freewheel rotates the brown flywheel. The output should have a smoother motion than the input.
    mechanism with freewheel.png
     
    Last edited: Jan 8, 2017
  10. Jan 8, 2017 #9

    JBA

    User Avatar

    One issue that may occur is that, even the flywheel oscillates at a high rate, the oscillating accelerations of the flywheel it may be difficult to deliver a smooth continuous motion to the bicycle driving wheel resulting in an effect similar to riding a single cylinder motorcycle at low rpm in which the sensation is a bit like riding a very short legged rabbit.
     
  11. Jan 8, 2017 #10
    That is one of my greatest concerns at the moment. However, I'm unsure of which flywheel you're referring too. I am aware that the purple flywheel intended to keep cadence will oscillate quickly. However, I'm hoping that the brown flywheel, intended to smooth out the motion, will not oscillate as you described.

    One method which I think may further smooth out the motion is to include two of these mechanisms on the crankshaft out of phase of one another, similar to multiple pistons on an engine. This is a rough outline:
    mechanism with double.png
    I am however hoping that with the blue freewheel and brown flywheel smoothing out the motion, this would be unnecessary.
     
    Last edited: Jan 8, 2017
  12. Jan 8, 2017 #11
    If this mechanism does work, there is something which I would like to know about it. (I did not ask it previously for fear of overcomplicating my initial post).

    How does the force applied on the crankshaft affect the rate at which the energy is released? These are two speculations I have:

    On one hand, a greater force applied should increase the acceleration of the flywheel. This increases its momentum. Therefore, it would take longer to change directions. This would mean an increase applied force would lead to a slower rate of energy release.

    On the other hand, the greater acceleration of the flywheel caused by an increased applied force should mean it would take less time to accelerate the flywheel after the direction changes. This would decrease the amount of time required. This would mean an increase applied force would lead to a greater rate of energy release.

    One idea I have is these two speculations would cancel each other out, meaning the force applied would have no effect on the rate in which energy is released. This would be like a pendulum which has the same frequency regardless of how hard it's pushed. However I have no real reason to believe that is the case.
     
    Last edited: Jan 8, 2017
  13. Jan 8, 2017 #12
    The first thing to do if you have an idea like this is to search " how much energy in coiled spring per Kg" .....this gives a figure of 2.3 WHr/Kg .... compare this to the energy stored in a Li-ion battery ...120 WHr/Kg !!

    So the idea of using springs to store energy is a complete non starter .....Don't lose heart , inventors usually have one good idea for every 100 bad ones.
     
  14. Jan 8, 2017 #13
    That does make sense. However, there is one other major difference between a spring and a battery of the same size- cost (not to mention the cost of a motor and generator). Although a bicycle powered by a battery is still the much better choice, one powered by a spring of the same size would be a lot cheaper to put together. It's probably not going to become successful in the market any time soon, but that isn't my objective at the moment- I just want to create designs and models for an interesting concept. If all else fails, it could be used for a toy.

    Also, where did you get 2.3 WHr/Kg- I couldn't find it when I searched " how much energy in coiled spring per Kg"?
     
  15. Jan 8, 2017 #14
    Since in this mechanism lot of sliding motion , you may expect higher frictional losses, which turn affects the overall efficiency of the mechanism
    To smooth out the speed fluctuations of flywheel, you have to take into consideration of mass moment of Inertia of the bicycle and rider
     
  16. Jan 8, 2017 #15
  17. Jan 8, 2017 #16
    I don't understand where sliding motion occurs in this mechanism. As I imagine it to work, no parts slide past one another.
     
  18. Jan 8, 2017 #17
    Between crank shaft and the bearings and between rack and pinion and between gear,pinion
    These are the points which will frictional forces
    Your aim is to smooth out the bicycle rider and not the flywheel
    So you ought consider the Inertia of the whole systems
     
  19. Jan 8, 2017 #18
    I don't see how this use of bearings would create more friction than normal. Technically, there would be friction from ball bearings. However, this would be the case for all mechanisms using them.
    As far as I can tell, the rack and pinion would only create sliding friction if the teeth fail to engage.
     
    Last edited: Jan 8, 2017
  20. Jan 9, 2017 #19
    Does anyone know the answer to this?
     
  21. Jan 10, 2017 #20

    Randy Beikmann

    User Avatar
    Gold Member

    I've got a few comments. I'm assuming that the spring on the crankshaft stores energy by being wound several revolutions like a clock spring, not just a fraction of a revolution.

    1) The crank/slider/rack/pinion will not affect the mean energy release rate of the wound spring. When the rack-mounted flywheel speeds up, it absorbs energy from the main flywheel, but when it slows sown, it returns the energy (minus friction). So if the main flywheel is rotating at, say, 300 RPM average, the crank/slider mechanism would just add an oscillation, so speed would vary from maybe 280 RPM to 320 RPM, repeatedly. And it will make the whole system shake, if it rotates fast.

    2) The rate of change of the main flywheel speed will depend on the storage spring torque (its stiffness times its deflection), the torque load driven by the shaft, the combined rotational inertia of all the moving parts, and any gear ratio between the spring and the load. Again, assuming no friction.

    3) Since the energy in a spring is proportional to the square of its deflection, energy release is nonlinear with rotation angle. Since the energy in a flywheel is proportional to the square of speed, the release of energy is nonlinear with speed change. This means it's difficult to control energy release rate without changing overall gear ratio as the system rotates.

    4) The ideal way to control the release of the stored energy would be to drive the load through an adjustable gear ratio (like a CVT). That would allow a wide range of "leverage" between the driving spring and the driven load.

    5) Every mechanical interface has sliding friction. For example, when gear teeth are engaged, they slide on each other as the gears rotate. The frictional loss is proportional to the torque on the gears. I would eliminate every possible mechanism, to avoid energy loss.

    6) Although a CVT has its own frictional loss, it would give you the most control of speed and energy release rate as the spring unwinds. By eliminating the other parts of your system, it might come out even.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted



Similar Discussions: Controlled Energy Release Mechanism
  1. Mechanical Speed Control (Replies: 17)

Loading...