What are those!!  

Re: What are those!!

I'm the one who made the add on feet to use sorbothane hemispheres or felt pads: https://www.thingiverse.com/thing:2802540/#comments - two versions of the feet, Fusion 360 files, and photos there.

With the sorbothane hemispheres I am experiencing the following:

I think mounting the printer as rigidly as possible and adding mass is going to be the best, but will only help so much. Any Y movement sends a lot of Z vibration into the machine. The Y axis is floating on two bars + belt + has bearings with slack that experience zero change when doing frame mods like these, and I think that is where much of the issue is.

How can we improve the rigidity and stability of the Y axis beyond the frame it is resting on?

Re: What are those!!

You can use screw driver to tighten or loosen the flexibility of the center anti-vibration core in those feet.

Looking forward to seeing them installed and hearing your opinion.

Re: What are those!!

Or just buy these https://www.ebay.co.uk/itm/263527552759

Re: What are those!!

I have noticed there is a fair bit of resonance. Really sends those vibrations through the frame. Tough thing to overcome.

Re: What are those!!

@bexwhitt, do those feet isolate vibration?

Re: What are those!!

My battle plan is to build an ikea enclosure, then mount the top of the frame to the inside of the enclosure. I did the same for my old monoprice select plus, it works well.

Re: What are those!!

@joshua, sounds interesting, do you have link to your enclosure or a build log?

Re: What are those!!

How can we improve the rigidity and stability of the Y axis beyond the frame it is resting on?
If Y motion causes the vibration problems on X/Z axis (which sounds credible as you describe it), then the logical step would be to reduce motion of the Y stationary part (motor, frame). Simplest way to do this is add mass to the areas where the force in Y direction is introduced into the frame, which is Y motor and Y idler. Again the simplest way to do this is to bolt the frame rigidly to a concrete paver or other super heavy stone plate (stone has higher internal damping than steel - old school would be cast iron). If you add a 20 kg plate to the 1 kg (assumption) Y frame, the Y travel as a result of the reaction force of the motor is reduced to 1/21. By "rigidly" I mean fixing it directly without rubber or felt pads. This way the mass of the stone base effectively reduces (as the accelerating force stays constant) the motion of the frame, which then (as you assume) introduces less ringing in the X/Z axis.

BTW nitpicking one might object that bolting down the frame is not optimal in terms of noise reduction: The rigid coupling also transduces motor vibrations onto the stone, acting as a loudspeaker. An optimal configuration would be a rubber coupling element that is stiff enough to couple the "baseband" Y motion to the stone directly, but decouples (and dampens) stepper motor noise cause by torque ripple (much higher frequency), so the mount crossover frequency should be between motion profile main spectrum and stepper ripple.

- Martin

Re: What are those!!

Very insightful, Thank you martin. I've seen people use hardwood plinths to isolate vibration from vinyl turntables, too

Re: What are those!!

Idea for DIY feet to stop vibrations

Re: What are those!!

Keep in mind that the goal is to dampen vibration, not isolate it. I use the iOS app "Vibration analysis" to identify the critical frequencies of the rig and then dampen each frequency with specific thicknesses of Sorbothane and other isolators.

Additionally, most vibration dampeners are designed to dampen external vibration in order to protect sensitive equipment. When using these, remember they must be inverted. For example, Sorbothane dampeners are best used upsidedown with the hemispherical side upward and the flat side down.

Our MK2s rig has 5 different dampeners, all tuned to specific frequencies.

Mark
@TOGOMedia

Re: What are those!!

One problem here is that the term "damping" is used with different meanings. It could mean vibration isolation, structural damping or vibration reduction. Each of those are implemented differently and might have quite contrary results. For example, if you mean vibration isolation (to reduce the noise the printer generates on a table), the most effective (not practical) way would be to hang it from the ceiling by a rubber string. This would eliminate coupling from the printer to the table, but maximize vibration transmission to the frame, as there is no counterweight to reduce it.

In engineering, the term "damping" is commonly used to identify a force that is proportional to velocity. Without damping, a mass together with a spring forms an oscillating system:
https://en.wikipedia.org/wiki/Vibration
In order to reduce the oscillation amplitude of such a system, you have to add damping. Several materials are commonly used, friction or oil dampers work, but rubber also works quite well. The amount of damping has to be tuned to maximize the effect. The downside of a damper is that it transmits a velocity proportional force, which is not a desired effect for vibration isolation.

So regarding your spring feet, the first question is what you are trying to achieve with them. You want to dampen... what exactly? My guess would be that the mode brigandier described further up sounds quite reasonable. He wrote Y motor motion (or the reaction force of the Y carriage movement) is transmitted to X/Z and causes structural vibration there, in addition to causing vibration on the main (extrusion+frame) assembly. As I already wrote the best method for this would be to reduce the Y motion by adding mass to it.
The spring feet have three issues:
1. By allowing more travel in Y direction, they increase vibration transmitted from Y to X/Z
2. Because they have no damping, they increase the structural vibration of the frame assembly
3. As they form an undamped oscillating system, they have the potential for resonance, https://www.google.ch/search?q=resonance+catastrophe&source=lnms&tbm=isch

So IMO they should make things worse. But who knows, I have not done a thorough analysis of this machine (and I am reluctant to do so, I'd rather spend time on 3D modeling than on debugging PRs design). Try it and post results.

If I had to bet money on where the ringing on X oriented faces would come from, my guess would be

50% as brigandier described (structural vibration excited by Y motor reaction force)
30% cogging of the linear ball bearings, together with low belt stiffness causing uneven tool motion (anyone use IGUS bearings?)
20% microstep nonlinearity (waiting for PR using the TMC2130 correction table, it's in the code already)

- Martin

Re: What are those!!

One problem here is that the term "damping" is used with different meanings. It could mean vibration isolation, structural damping or vibration reduction. Each of those are implemented differently and might have quite contrary results. For example, if you mean vibration isolation (to reduce the noise the printer generates on a table), the most effective (not practical) way would be to hang it from the ceiling by a rubber string. This would eliminate coupling from the printer to the table, but maximize vibration transmission to the frame, as there is no counterweight to reduce it.

In engineering, the term "damping" is commonly used to identify a force that is proportional to velocity. Without damping, a mass together with a spring forms an oscillating system:
https://en.wikipedia.org/wiki/Vibration
In order to reduce the oscillation amplitude of such a system, you have to add damping. Several materials are commonly used, friction or oil dampers work, but rubber also works quite well. The amount of damping has to be tuned to maximize the effect. The downside of a damper is that it transmits a velocity proportional force, which is not a desired effect for vibration isolation.

So regarding your spring feet, the first question is what you are trying to achieve with them. You want to dampen... what exactly? My guess would be that the mode brigandier described further up sounds quite reasonable. He wrote Y motor motion (or the reaction force of the Y carriage movement) is transmitted to X/Z and causes structural vibration there, in addition to causing vibration on the main (extrusion+frame) assembly. As I already wrote the best method for this would be to reduce the Y motion by adding mass to it.
The spring feet have three issues:
1. By allowing more travel in Y direction, they increase vibration transmitted from Y to X/Z
2. Because they have no damping, they increase the structural vibration of the frame assembly
3. As they form an undamped oscillating system, they have the potential for resonance, https://www.google.ch/search?q=resonance+catastrophe&source=lnms&tbm=isch

So IMO they should make things worse. But who knows, I have not done a thorough analysis of this machine (and I am reluctant to do so, I'd rather spend time on 3D modeling than on debugging PRs design). Try it and post results.

If I had to bet money on where the ringing on X oriented faces would come from, my guess would be

50% as brigandier described (structural vibration excited by Y motor reaction force)
30% cogging of the linear ball bearings, together with low belt stiffness causing uneven tool motion (anyone use IGUS bearings?)
20% microstep nonlinearity (waiting for PR using the TMC2130 correction table, it's in the code already)

- Martin

This. Some of the solutions in this thread will likely make the problem worse.

My approach (to reduce frame oscillations, not noise) is to tighten things up which pushes the natural resonance up above most of the input excitations. The hard feet (see my sig link) that I use on a paver stone do this well and has a bonus side effect of also quieting the system. I plan to reinforce the Z frame which will complete that effort.

Re: What are those!!

50% as brigandier described (structural vibration excited by Y motor reaction force)
30% cogging of the linear ball bearings, together with low belt stiffness causing uneven tool motion (anyone use IGUS bearings?)
20% microstep nonlinearity (waiting for PR using the TMC2130 correction table, it's in the code already)

- Martin

I am using IGUS in my Y axis, and they did help a bit. Unfortunately, one of my Y axis stock bearings was bad and grinding a bit, so I can't say if my improvement would be the same for everyone. I used them in X for a bit too and swapped back, the X carriage squeezes them too hard and deforms them.

In regards to the Y axis dumping vibration into X/Z, my biggest improvement in that area was sorbothane feet. Compared to the rubber stock feet, sorbothane acts differently by (if I recall correctly) dumping vibration out at different angles and converting much of it to heat. However, this improvement wasn't what I expected. The problem is still there with the same intensity, but length of the artifact it creates is pretty much half.

To really drive home that change, if I had 20mm of heavy ringing after turning a corner, now I have 10mm~.

This makes me think Martin is correct. The sorbothane helps because it is dumping the movement in other directions and as heat; however, they can't do it instantly. They just do it a hell of a lot better than rubber. Increasing mass or making that axis rigid sounds like more of a fix.

Re: What are those!!

Increasing mass or making that axis rigid sounds like more of a fix.
Do you already have a paver? My guess is that the harder you fix the printer frame to the paver, the less motion you'll have transmitted to X/Z.
Look here: http://www.mikro.co.kr/eng/bbs/skin/eng_product/popup_view_image.php?tb=hbbs_eng_product&no=29&no2=0
This is how it is done industrially (I spend quite some time debugging the machine that goes on that). The mineral cast base has inserts where the axis drives are fixed to. And look where the rubber feet are.

I'd print some hard feet (four at the ends of the extrusions and two in the middle for eliminating the bending mode) and fix them somehow (hot glue?) to a heavy stone block. I have the paver on an inflated bike tube, but this is for isolation as I want to run the printer over night and I can't have the noise.

- Martin

Re: What are those!!

I found these see these on that site, https://mnpctech.com/pc-computer-stereo-desktop-case-feet/3d-printer-base-feet/

Re: What are those!!

@bexwhitt, do those feet isolate vibration?

They seem to do, the screws are below the surface, and the rubber is of a similar stiffness as the ones supplied without the coming detached.

Re: What are those!!

@bexwhitt, do those feet isolate vibration?

30'30 tee-nuts are not as common as the smaller sizes you could easily discard the rubber bits and print something better, the best way to counter vibration BTW is a paving slab.

Re: What are those!!

I went the LACK enclosure route... and it was shit... for stability and vibration. It's probably not bad if you have it sitting on a table and have a LACK just covering it... but double stacking a LACK and putting a printer on them, is no good IMHO.

best bet is to start with a table that is level with all feet squarely on the ground and go from there.

my .02

Andy