Speed of Springs?

It's probably easier to think in terms of a mass being moved linearly by a compression spring (the idea is the same, just not rotary). So, a compressed spring would accelerate the mass, though not with a constant force (acceleration) to the formerly relaxed position where the speed would be a maximum. At that point most of the energy of the spring would have been delivered to the mass. At that point, you could theoretically transfer the mass to a shorter, stiffer spring that would further accelerate the mass (matching speeds would be tricky, but theoretically possible). And you could repeat this several times, though each time you would probably lose an increasing amount of energy in the transfer. But, you would quickly run into a physical limitation of how short and stiff you could make the next spring.

So, yes, it's possible, but only for a short spin-up. This sort of problem has been more or less successfully handled with electrical pulses. Is that a possibility?

I believe I would consider the use of 4 or more springs for the loading so the pressure on the end spring would not wear it out fast or temper it with constant use. Plus the use of several springs should give you more adjustment for the release of engery. You know the release of engery will depend on the spring and the material it is made of.

I'm guessing it's a fairly small/lightweight flywheel? Otherwise, I think you'll run short of elbow grease too early to reach 1K rpm's.

can you make your flywheel into a magneto? magnetize the flywheel and add some brushes and you have a generator...

Thank you very much to all who responded so promptly; TVP45, dadw5boys, EnviroMan, and JEP. This is the one and only time I've used this forum and I'm impressed by how it works.

The idea of converting a rotary problem to a linear one is a valuable aid to getting a picture of the situation. Thanks TVP45. Clearly, the energy in the spring must accelerate its own mass as well as the mass of the flywheel and the spring force, as you point out, will decline as the spring unwinds toward its formerly relaxed position.

dadw5boys, I envisage driving the flywheel through a freewheel (as in a bicycle), so that when the spring is released from a loaded position into the freewheel, the freewheel will engage almost immediately and the spring will apply torque through it to the flywheel. This is of course, provided that the spring is unwinding at a rotation speed faster than the flywheel. (If the spring unwinds at a slower speed than the flywheel, then the freewheel will not engage and the spring will simply unwind and presumably its energy will be simply converted into heat.)

EnviroMan. Good point. Yes, my own elbow grease is definitely limited. On the other hand, if the spring speed is sufficient and the flywheel drag suitably minimized, presumably the flywheel speed can be increased at a slower rate toward the target rate.

Thanks for the suggestion, JEP.

Edgy Ed

Do you have to have the spring as an intermediate storage stage. My son's got one of these wind up and release toys:-

http://www.amazon.co.uk/Hot-Wheels-Power-Rev-Motobike/dp/B000SSL5NU/ref=sr_1_9?ie=UTF8&s=kids&qid=1219044163&sr=1-9

Just requires good gearing and clutch/ratchet mechanism.

Good thought. A ratchet would make all the difference, I think.

I believe you will find the velocity of the spring is a sine function, and is equal to the resonant frequency of the free spring, in its unloaded condition, or the spring plus any attached mass, in its loaded condition.

080818

Hello.

Thank you all who commented further, Randall, TVP, Welderman.

Yes, it is necessary to have the spring as an intermediate storage stage. The advantage of the spring in the energy flow is that a spring can be loaded at a wide variety of discontinuous input rates, as input energy becomes available (depending, for example, limited elbow grease, coffee breaks, time out to watch the game or ...) and released at a reasonably fast rate (hopefully, fast enough to drive a flywheel to, say, 1000 rpm). (Consider an ordinary spring-loaded tape measure. The tape can be pulled at any discontinuous rate and kept out for as long as desired. But when it's released, it goes like stink.)

For this conceived application, the spring could be loaded using a suitable ratchet mechanism and after loading would be released as required, using a suitable clutch/release mechanism, through a freewheel, which will transmit energy from the spring to a flywheel. Clearly, energy transfer is possible only if the spring is moving faster than the flywheel, and can transmit energy at the flywheel's speed. My objects for this project include having a simple system, with only mechanical energy input, without gearing, able to absorb a lot of physical abuse, ... and at least modestly efficient ( in terms of the input energy which is eventually transferred to the flywheel).

Edgy Ed

And, using your example of the tape measure, the spring can be pre-loaded before you need the conversion of potential energy to motion.

It seems to me, that if you obtain a wind-up clock movement, you will have the device you ar looking for. You can wind the clock spring at any desired rate, and use the stored energy to drive a flywheel instead of the clock escapement.

080819

Hello. Thanks for your comments, EnviroMan and welderman.

Yes, your comments are right on, certainly in line with my thinking. However, I really need some numbers to crunch (ie. how fast? how much energy transfer? ...) so that I can save myself grief in building a real machine. A useful substitute for numbers would be more examples of fast energy/motion transfer from a spring which would give me more to work with. A wind-up clock mechanism is a good example but I am unclear about its actual attainable speed.

Edgy Ed

Check out the differences/sameness with those hand-cranked weather radios and electric lanterns.

The unwinding speed depends on the connected inertia. If you consider the linear spring-mass the equivalent inertia mass of the spring is 1/3 of its own mass. As long as the added mass is significant with respect to th own spring mass the spring alone will unwind faster. So that your question is answered. The problem with a torsional spring is similar. With respect to the possible speed you should consider the nergy required to bring the flywheel to 1000rpm and make sure that the spring in its loaded status will accumulate this amount to which you should add ALL possible friction losses:

-bearings (better ball since those have the lowest internal roll friction)

-air (if the flywheel is thinn and has a big diameter)

- transmission

a. s. o.

With usually on internet available documents you can solve with an uncertainty of 10 to 15% your problem before making any prototype.

I suggest you google "flywheel losses". You will find a wealth of information regarding friction, inertia, aero-loss, etc.

The spring self inertia (for torsion) or self mass (for compression/tension) need to be considered. A good approximation is to take 1/3 of the spring mass or inertia.