Solenoid driven pendulum

[Ted Ewert]

This experiment is one I’ve been working on for the last few days. I’m exploring different ways of driving a pendulum with a solenoid.
I’ve developed a solenoid which uses magnets within the steel to add power to the armature. It can generate over 100 lbs of pull so I wanted to find a way to put it to good use. Segmented armature motor - Heretical Builders
I’ve tried a number of configurations. What I have found is that a solenoid doesn’t always have the kind of power curve which is required in a given application. While powering the pendulum by hand seems relatively simple, duplicating the human power curve with the solenoid is another matter.
Nevertheless, after a few days of testing I’m starting to get the hang of it. I salvaged a clutch bearing (one way bearing) off an old Nordic Trak exercise machine. I attached a sprocket to the end of it so I could drive it with a chain. This makes an excellent drive mechanism for converting linear motion into angular motion.

The limiting factor with a solenoid is it’s stroke length. The pull applied to the armature increases exponentially as the stroke gets shorter. It also decreases exponentially as the stroke length increases.
I have found that increasing the stroke length helps to match the effective power curve of the configuration. Even though there is little torque at the beginning of the stroke, a more gradual torque buildup works much better. This may also have to do with the clutch bearing grabbing better, but I’m not sure about that.
Here is a video I did on an earlier version.
YouTube - Solenoid Powered Pendulum
This Milkovic type configuration involves the solenoid pulling the lever arm down to indirectly add energy to the pendulum.



It was technically successful but far from efficient. Much more work needs to be done to effectively match the power curve to an oscillator type mechanism like the two stage. If this were accomplished, I believe the solenoid could be a very effective drive mechanism.
I went from that to a more direct drive approach. This is where I started using the clutch bearing.
YouTube - Solenoid powered pendulum 2.wmv


This drove the pendulum much better once I lengthened the stroke and the duty cycle. (Don't ask me why the clamps are there... )
I'm going to build a new type of "solenoid" since I think this design has some inherent weaknesses. It won't technically be a solenoid, but will do the same thing. Stay tuned...

Cheers,

Ted

[cody]

Cool, something about things that move that put a smile on my face.


One thing that came to mind about the power curve not being right is to use your senser to trigger a function generator that would then output a ramp signal to your transistor that powers the coil. Or even program the exact curve into the generator if its capable.

[rave154]

Ted,

nice mods !!!

that's a nice "elegant" solution to the "linear-to-rotational" problem there, also any pull generated by the solnoid tranlates to instant torque on the rotational shaft.

may i ask, are you planning to put generator / pickup coils & neos on the rotating arm? or similar?

or are you planning to actually drive something rotationally? ie:- a car, a pump etc?

a thought occured ot me watching your video there...

if theres no actual need for roational motion as an output... then the swing pendulum arm could be removed...and simply have the solnoid pulling in a straight line.on the chain..with the other end of the chain attatched to a suitable spring for return.... neo's mounted on the chain..with appropriate pickup coils surrounding the chain along its length... something like that.

on that note... assuming you have a very strong solonoid, but it has a small travel...say.... 1cm or so.... what arrangement of neos & pickup coils would be best to translate this to electrical power... as in.... strong neos & very thick windings...or lots of turns of medium windings?


look forward to your answers and more mods

[Ted Ewert]

Cool, something about things that move that put a smile on my face.


One thing that came to mind about the power curve not being right is to use your senser to trigger a function generator that would then output a ramp signal to your transistor that powers the coil. Or even program the exact curve into the generator if its capable.

Hi Cody,
I like mechanical stuff too, much easier for my small brain to understand.
Modifying the power curve is easier done mechanically than electrically. Adjusting the voltage is about as much electrical modification as I care to have.
Besides, I'm just trying to get a general idea of how efficient different configurations are at this point. Thanks for the suggestion though.

Cheers,

Ted

[Ted Ewert]

Ted,

nice mods !!!

that's a nice "elegant" solution to the "linear-to-rotational" problem there, also any pull generated by the solnoid tranlates to instant torque on the rotational shaft.

may i ask, are you planning to put generator / pickup coils & neos on the rotating arm? or similar?

or are you planning to actually drive something rotationally? ie:- a car, a pump etc?

a thought occured ot me watching your video there...

if theres no actual need for roational motion as an output... then the swing pendulum arm could be removed...and simply have the solnoid pulling in a straight line.on the chain..with the other end of the chain attatched to a suitable spring for return.... neo's mounted on the chain..with appropriate pickup coils surrounding the chain along its length... something like that.

on that note... assuming you have a very strong solonoid, but it has a small travel...say.... 1cm or so.... what arrangement of neos & pickup coils would be best to translate this to electrical power... as in.... strong neos & very thick windings...or lots of turns of medium windings?


look forward to your answers and more mods

I've thought quite a bit about what you're suggesting there with a linear generator of some sort. It's on the list, but first I want to improve the motor part.
While walking the dogs this morning I got to thinking about the present configuration I'm using for the solenoid, and I realized a serious flaw in my original theory. Consequently, I'll be building another type of linear motor which I think will be much stronger for the same power input.
Two opposed linear motors in a sequential pull/pull configuration could run a linear generator quite well. I've already got the plan ready to go in my head.
I really like Dragon's linear generator (below), and would probably build something similar.


Cheers,

Ted