Springs with the least amount of damping?

[Rhettoric]

I have a project I'm tinkering with that involves a "gyroscope" that I want to rotate back and forth using a spring. The gyroscope is a piece of steel pipe with a 4in ID about a foot long that weighs roughly 10lbs or so. I have a 5/16ths rod going down the center as an axle and it's running on skateboard bearings. The gyroscope seems to work pretty well at retaining it's rotational energy when spun.

I'm having trouble finding a way to make it rotate back and forth with a spring though. I've tried many different types but none of them give anywhere near the performance I was hoping for. I understand there will always be losses, but I feel I'm getting way too much.

I've been running a test where I tie a string to the axle near one of the bearings (to reduce any possible deformation to the axle) and anchor the string to a spring on my workbench. Then I rotate the gyroscope 360 degrees and let go to observe how long it will oscillate for.

No matter what type of spring I'm using, the gyroscope seems to lose about 20-30% of it's rotation every time it switches direction, coming to a relative standstill in rather quickly.

Seeing as the gyroscope itself is rather efficient at retaining rotational energy, I decided to test just the springs themselves by hanging a weight from them and seeing how long they would maintain a vertical oscillations. The results weren't much, if any, better which leads me to believe the springs are the primary cause of the observed losses.

Given that I need the rotation to switch direction relatively quickly (no more than about a second or so) I'm going to need some pretty beefy springs, but I've tested weaker ones also and they seem to be just as inefficient.

Does anyone know what types of springs have the least damping, or happen to know of another way to approach this problem? I've tried googling for several days, but most of the results for "efficient springs" and similar searches come up with stuff about spring cleaning and other stuff not even close to what I'm looking for :)

Thanks for any help or info you can offer!

Rhett

 

[AlephZero]

You don't say what you hoped to achieve, but the best you are likely to get, including the damping effect of the air, would be a modal damping coefficient of about 0.02.

That will lose about 10% of its amplitude for each complete oscillation, and about half the amplitude in 6 or 7 oscillations.

If you want to do better than that, the best "simple" improvement would probably be to run the system in a vacuum chamber to get rid of the air damping. The next biggest source of damping will probably be friction in any mechanical joints in the system, including how the device is clamped to the workbench. The internal damping inside metal components like the spring (I'm assuming the spring is metal) is probably not worth bothering about until you get rid of those issues.