I have a feedback problem between my old turntable and speakers. I'm not crazy about using multi-pound weights, a box full of sand or an ugly-as-sin inner tube to address the problem so I started experimenting with a different approach. It started with some floor protectors that are designed for things that have casters on their feet -- they have a cylindrical depression in them, so a small ball bearing can easily roll back & forth inside them For my first experiment I actually got some 1/2" silicone balls, on the theory that they could also damp a certain amount of vibration in the vertical plane. The protectors were placed, plastic spheres dropped in, a plywood piece on top, and finally the TT went down.
It worked pretty darned good! I thought things could be further improved because the cylindrical "bowl" permits the spheres to freely move along the axis of the cylinder so the TT had a tendency to drift around. What I really needed was some "pucks" with a spherical bowl shape so they would work on both the X and Y axes. I looked around but didn't find anything the right size, or didn't cost too much -- so I decided to make some.
Here's my first shot at a home-made version:
I have a small machine shop so it actually wasn't too difficult to make this guy. I first tried printing one on my 3D printer but the vertical steps interfered with the free motion of the sphere. It probably would work with something better than an Ender 3. An SLA printer might be the ticket, but I don't have one.
Interestingly, the oscillation frequency of the sphere is independent of the TT mass, it only depends on the bowl's radius of curvature. And it doesn't take a very large radius of curvature to achieve a very low effective cutoff frequency. I calculate that a bowl with a 4.4cm radius of curvature has an oscillation frequency of just TWO HERTZ!
The one in the photo has a natural oscillation frequency around 3 Hz so I believe the math. I'm grinning a lot over this one.
The fact that the frequency doesn't depend on the TT mass resting on them means that the entire system, consisting of 4 of these pucks, has the SAME cutoff frequency as just one of them. I love physics....when it works in my favor 😆
I had thought that the oscillation frequency would be the same as for a pendulum, but it isn't. The fact that the sphere rotates (due to friction) changes its natural oscillation frequency so the oscillation period is slightly longer, compared to a simple pendulum. The approximate period is given by: T = 2*pi*sqrt((7R)/(5g)), where R is the bowl's radius of curvature and g is the gravitational constant. SI units -- R is in meters, g is 9.81 m/s^2. T is seconds, of course -- 1/sqrt(s^2) = s. I found the formula over on physics.stackexchange.com. It's valid for small-amplitude oscillations, which should be the case here. A simple pendulum's period is given by T = 2*pi*sqrt(R/g) so T isn't THAT different, just a bit longer. Longer is better.....
The shallow bowl means that it wouldn't be too difficult to push the TT off of them, so my final puck design will include a 3D-printed skirt around the outside of each puck so the sphere will fetch up against it and stop.
I bought the spheres from McMaster-Carrr so I could get them quickly, but that ended up costing me $35 for 10 of them. Rubber spheres can be had on ebay or Amazon for less, if you're willing to wait. Grainger might be another source. I got the acetal plastic from Tap Plastics, $5 for 4 of them, 1/2" thick. There also are online metal and plastic suppliers that will cut material to size, although they usually charge a small fee for special cuts. If you go that route, shop around first -- fees can vary quite a bit.
I realize that few of you will own, or have access to, the machine tools I used to make these, but there are other ways to make something similar. For instance, it wouldn't be too difficult to make a mold using a large ball, to cast them out of epoxy resin. The inside surface of a large ~2" ABS plumbing fitting -- an end cap -- has a decent radius. But Its exterior also has a radius, so I don't know how that will work -- if left free to rock, it might behave like a flat piece & that's not what you want. But it MIGHT make be good for creating the bowl, maybe pressing it into epoxy putty. Put mold release on it first.....
It worked pretty darned good! I thought things could be further improved because the cylindrical "bowl" permits the spheres to freely move along the axis of the cylinder so the TT had a tendency to drift around. What I really needed was some "pucks" with a spherical bowl shape so they would work on both the X and Y axes. I looked around but didn't find anything the right size, or didn't cost too much -- so I decided to make some.
Here's my first shot at a home-made version:
I have a small machine shop so it actually wasn't too difficult to make this guy. I first tried printing one on my 3D printer but the vertical steps interfered with the free motion of the sphere. It probably would work with something better than an Ender 3. An SLA printer might be the ticket, but I don't have one.
Interestingly, the oscillation frequency of the sphere is independent of the TT mass, it only depends on the bowl's radius of curvature. And it doesn't take a very large radius of curvature to achieve a very low effective cutoff frequency. I calculate that a bowl with a 4.4cm radius of curvature has an oscillation frequency of just TWO HERTZ!
The one in the photo has a natural oscillation frequency around 3 Hz so I believe the math. I'm grinning a lot over this one.
The fact that the frequency doesn't depend on the TT mass resting on them means that the entire system, consisting of 4 of these pucks, has the SAME cutoff frequency as just one of them. I love physics....when it works in my favor 😆
I had thought that the oscillation frequency would be the same as for a pendulum, but it isn't. The fact that the sphere rotates (due to friction) changes its natural oscillation frequency so the oscillation period is slightly longer, compared to a simple pendulum. The approximate period is given by: T = 2*pi*sqrt((7R)/(5g)), where R is the bowl's radius of curvature and g is the gravitational constant. SI units -- R is in meters, g is 9.81 m/s^2. T is seconds, of course -- 1/sqrt(s^2) = s. I found the formula over on physics.stackexchange.com. It's valid for small-amplitude oscillations, which should be the case here. A simple pendulum's period is given by T = 2*pi*sqrt(R/g) so T isn't THAT different, just a bit longer. Longer is better.....
The shallow bowl means that it wouldn't be too difficult to push the TT off of them, so my final puck design will include a 3D-printed skirt around the outside of each puck so the sphere will fetch up against it and stop.
I bought the spheres from McMaster-Carrr so I could get them quickly, but that ended up costing me $35 for 10 of them. Rubber spheres can be had on ebay or Amazon for less, if you're willing to wait. Grainger might be another source. I got the acetal plastic from Tap Plastics, $5 for 4 of them, 1/2" thick. There also are online metal and plastic suppliers that will cut material to size, although they usually charge a small fee for special cuts. If you go that route, shop around first -- fees can vary quite a bit.
I realize that few of you will own, or have access to, the machine tools I used to make these, but there are other ways to make something similar. For instance, it wouldn't be too difficult to make a mold using a large ball, to cast them out of epoxy resin. The inside surface of a large ~2" ABS plumbing fitting -- an end cap -- has a decent radius. But Its exterior also has a radius, so I don't know how that will work -- if left free to rock, it might behave like a flat piece & that's not what you want. But it MIGHT make be good for creating the bowl, maybe pressing it into epoxy putty. Put mold release on it first.....
I finished making 4 of these things and had a chance to try the setup. It seems to work pretty good -- I'm not hearing any evidence of feedback.
The silicone spheres I'm using are a bit on the soft side so the TT lists a bit to one side. I ordered some harder nitrile spheres to see if that helps. If need be I can add some shims to the low side.
One potential problem is so-called "set" -- some kinds of rubber will slowly deform under pressure, so in this application the spheres could develop flat spots that would inhibit their rolling action. Silicone rubber is one of the best in terms of lowest percentage of set. The spheres I got have a hardness rating of 35, on the soft side. The harder nitrile spheres probably will be a bit worse in terms of set. I found some harder silicone rubber spheres but will hold off on buying them for now. I'll give the nitrile ones a chance first, maybe they will be good enough for my purposes. Problem is, they are on a slow boat from China so I won't get them until mid-June. In the meantime I will make myself some shims. Who knows, maybe the nitrile spheres will go into my growing supply of "I'll get around to this later" items......
Steel spheres would be the absolute best, no set whatsoever -- but they wouldn't absorb any vibration in the vertical plane. They also would transmit any imperfection in the two surfaces -- my puck and plywood support -- as they roll. And ZERO damping. I have serious doubts w/regard to them.
The silicone spheres I'm using are a bit on the soft side so the TT lists a bit to one side. I ordered some harder nitrile spheres to see if that helps. If need be I can add some shims to the low side.
One potential problem is so-called "set" -- some kinds of rubber will slowly deform under pressure, so in this application the spheres could develop flat spots that would inhibit their rolling action. Silicone rubber is one of the best in terms of lowest percentage of set. The spheres I got have a hardness rating of 35, on the soft side. The harder nitrile spheres probably will be a bit worse in terms of set. I found some harder silicone rubber spheres but will hold off on buying them for now. I'll give the nitrile ones a chance first, maybe they will be good enough for my purposes. Problem is, they are on a slow boat from China so I won't get them until mid-June. In the meantime I will make myself some shims. Who knows, maybe the nitrile spheres will go into my growing supply of "I'll get around to this later" items......
Steel spheres would be the absolute best, no set whatsoever -- but they wouldn't absorb any vibration in the vertical plane. They also would transmit any imperfection in the two surfaces -- my puck and plywood support -- as they roll. And ZERO damping. I have serious doubts w/regard to them.