Best material to reduce those pesky vibrations

Hello. I just completed a CoreXY build that uses NEMA 23's everywhere but the extruder. The motors run fine but they have a lot of vibration that gets put directly into the frame and make any movement very loud. I am looking for some materials people have had success with that dampen these vibrations. I have seen cork and these Astrosyn dampers. Also, I came across this design and was wondering if it could be applicable at the 3D printing scale. Any help would be appreciated.

Thanks
JTT

You do not want any flywheel effect on your steppers. The best way to dampen is to isolate the steppers from the frame, the astrosyn dampeners work well enough. If your still getting allot of vibration on the frame then your probably looking for noise dampening feet for the printer.

---

My Personal Blog. Build blog.
[engineerd3d.ddns.net]

Modicum V1 sold on e-bay user jaguarking11

Actually, since I have installed the Gadget3D stepper drivers into my printer, the vibrations are completely gone!

Not sure if those are fit for Nema23 steppers, but I am going to buy them for my other printer too.

There are special-cut pieces of cork available from various sellers, but you can also just try to make it yourself by using a rubber mat and cutting it up in the right dimensions. I have seen others use those solutions, but personally didn't feel like it really dampened much.

I must admit that my CoreXY printer is way more vulnerable to the vibrations you mention, it vibrates way more than my Air 2 or my Mendelmax 1.5

Edited 1 time(s). Last edit at 02/02/2015 01:25PM by Ohmarinus.

---

http://www.marinusdebeer.nl/

I just modeled up some "DIY" Astrosyns that use mounts in your material choice and Ninjaflex(or other flexible material) glued in the center. I made these for NEMA17 motors but fairly simple to resize it for 23s if it works out, should also be able to adjust the dampening with material/perimeters/infill. Will test tonight and upload to Thingiverse and Youmagine.

---

Please add your printer design to [reprap.org]

I made something similar as tjb1, you can find the files here and modify to fit the nema 23:
[github.com] or here: [www.youmagine.com]

Anyway, the Astrosyn dampers are the best, but even if you use the printable ones, the difference will be so much noticeable.

edit 3 minutes later: if you want, I can update the design for nema 23s and if space allows, it's best to print them way thicker.

Edited 1 time(s). Last edit at 02/02/2015 03:35PM by imyz.

Thanks for the quick responses guys. I currently have my entire printer sitting on a 1" thick sheet of styrofoam. This worked very well for quieting my last printer but doesn't seem to have much of an effect on this one. Before putting the motors on the printer I set them on a piece of this styrofoam and tested the noise level compared to them sitting directly on the table the printer is on. There was a big difference. Almost no sound at all on the foam. Would dense rubber feet work better for this? Also, when isolating the motors from the frame is a dense material like rubber better for high frequencies or just good all around? What about a thin sheet of foam between the motor and frame?

JTT

I use several layers of the non slip matting under mine. You need a few layers, but it kills most noise.

Here is mine, feels ok but I am not ready to tear printer apart yet to test it.

[www.thingiverse.com]

---

Please add your printer design to [reprap.org]

Quote
jtoombs
Also, I came across this design and was wondering if it could be applicable at the 3D printing scale.

Wow, I'm compelled to comment on how misinformed that guy is regarding harmonic dampers. He does not understand the purpose, method of operation, or proper use of a harmonic damper, and has no idea what causes a stepper motor to make noise. He is literally planning to glue a rattle to the back of his stepper motor because he thinks it will make it quieter.

Just so everyone is aware, stepper motor noise is almost exclusively caused by poor drive electronics which lead to resonance (and coil whine). If you measure shaft position over time, the single-chip stepper drivers tend to look like this:


If you are willing to pay 40$ for an anti-resonance digital stepper driver your motor will be completely silent. Microstepping can reduce resonance as well, but most single-chip drivers switch in the audible range so there will always be coil whine.

Edited 2 time(s). Last edit at 02/03/2015 12:11AM by 691175002.

Just ordered a box of Astrosyn for Nema 17's, so I'll do a little video once I get them so people can see (hear) any difference.

Hey guys. Thanks for those links to the printed Astrosyns. Tjb1 and Imyz, I would really appreciate it if you could post a NEMA 23 sized part on Thingiverse etc. I bought a sheet of anti-vibration rubber called sorbothane. I don't know if anyone has used it but I will post results when I get it. Maybe I could put it in between the two separated plates of the Astrosyn style dampers.

Regarding the design 691175002 commented on: After reading and watching more videos on this concept I think it is more of some sort of an inertial mechanism that helps keep CNC motors from stalling and making racket (This is just a speculation. I don't actually know for sure...). But it does seem to reduce the noise at the higher speeds CNC machines need to drive the leadscrews at reasonably quick speeds. At lower speeds, though, it is basically a rattle.

JTT

May I asked about the reason of Nema 23 selection ? Sometimes bigger does not means better. You experienced that Nema 23 will vibrate much more than nema 17 at low speed. And on our printers, the speed of our motors is not that high. The Nema 23 runs better at higher RPM. Reducing vibrations can be done with better drivers and/or higher resolution microsteping. Maybe it's wise and cheaper to go back to Nema 17, even powerful ones. Otherhand, the only way to kill vibrations is to absorb them through mass.

---

Collective intelligence emerges when a group of people work together effectively. Prusa i3 Folger (A lot of the parts are wrong, boring !)

I have tried the Astrosyn dampers on the NEMA-23 motors in my printer and they made no discernible difference in the vibration and noise. They will help prevent vibrations coupled from the body of the motor to the machine frame, but in my case, and probably yours, the noise from the frame is minimal. The vibration in the rotor will still couple to whatever the rotor connects to.

I also use DSP based drivers, which helped just a little even though I am running 64:1 microstepping.

I have just printed a harmonic damper of the type referenced by the original poster and I'm testing it to see if it helps with the very objectionable noise I get on the ball screw driven Y-axis of my printer. The bed is a 12.5 x 12 x 1/4" aluminum plate suspended by three screws at the edges of the plate essentially making it a bell. And it rings terribly... I'm hoping that damping the vibrations in the rotor with the harmonic damper will reduce the ringing from the bed plate.

If anyone is interested I can supply an stl file for a 100% printable harmonic damper that is designed to hold 25 pennies and fits onto a flatted 1/4" shaft commonly found on dual shaft NEMA-23 motors.

Edited 1 time(s). Last edit at 02/03/2015 09:02PM by the_digital_dentist.

Zavashier, I am using NEMA 23's because I got them for free. Plain and simple. Four of the same type of motor in great condition. They were just going to be thrown out until my keen eye for printer parts spotted them. So, naturally, I thought I would give it a try. I do realize that the bigger motor doesn't really benefit me in any way at this scale but like I said I wanted to at least try. Also, I have already printed all the parts for this printer using a now dismantled printer that I needed to get some frame materials out of for my current printer (kind of a cannibal printer situation)... On a different note, I did manage to reduce some noise by properly tuning the pots on my stepper drivers using the Vref method. This has not gotten rid of it completely but hopefully the sound dampening material I purchased will give some good results.

Dentist, from your other posts I have seen around the forums you have leadscrews on all of axes, right? I hope this harmonic damper produces good results for your printer. It seems like it has good potential due to the similarities between your printer and CNC machines which this device is used on. Let me know what happens when you complete your tests. Unfortunately, my motors are not dual shafted. Not the optimal situation for a simple harmonic damper. I would still like to see your design if you don't mind posting it.

JTT

Just saw this on RobotDIGG:
[www.robotdigg.com]

I like their shop, it's no secret, you can email them and ask if they can also supply you with Nema23 dampeners. They are very flexible with customs things, however, in some cases they ask for a M.O.Q.

---

http://www.marinusdebeer.nl/

Those are essentially the same as the astrosyn dampers that don't help much at all.

Quote
jtoombs
Dentist, from your other posts I have seen around the forums you have leadscrews on all of axes, right? I hope this harmonic damper produces good results for your printer. It seems like it has good potential due to the similarities between your printer and CNC machines which this device is used on. Let me know what happens when you complete your tests. Unfortunately, my motors are not dual shafted. Not the optimal situation for a simple harmonic damper. I would still like to see your design if you don't mind posting it.

JTT

I just have screw drive on the Y and Z axes. The X axis is very quiet, especially after I added the DSP driver.

My initial test with the harmonic damper looks promising. I just left the pennies loose in their holes. I'm going to glue them together and try again. It was a bit rattly with them loose, but it did seem to help a lot with the terrible noise the printer was producing- my wife didn't complain once when I was running the printer with the damper attached to the Y axis. I'll try to run another test tonight and make some video of the results.

Edited 2 time(s). Last edit at 02/04/2015 05:52PM by the_digital_dentist.

OK, testing done and video uploading to vimeo. Result: the Y axis screw drive seems as noisy with the harmonic damper as without. The harmonic damper rattles and adds to the awful noise the printer makes so it is actually worse with it than without it.

Video here: [vimeo.com] and here: [vimeo.com]

Design here: [www.youmagine.com]

The harmonic damper demos by the CNC machine people that show miraculous results are for fast moves in one direction, not a lot of low speed start, stop, and reverse that a 3D printer does (or a CNC mill when it's actually cutting).


Moving on...

I think I need to start looking into a different print bed design that isn't a bell/cymbal that is free to rattle so much.

Edited 2 time(s). Last edit at 02/04/2015 11:36PM by the_digital_dentist.

Looks like the links to your videos don't work. I think the added challenge of making it a level-able bed is what makes 3D printer beds so prone to rattling. That's why I wanted to have the bed only move in the Z axis. Anyway... I will get my sound dampening sorbothane tomorrow and will hopefully have results to post by tomorrow night.

JTT

Quote
the_digital_dentist
Those are essentially the same as the astrosyn dampers that don't help much at all.

I got my Astrosyns today to have a play with. First thing that I should of spotted when I looked at them on the web site, but didn't, is they all expect you to be mounting the motor using two screws across the diagonal. So the standard mount for the Y motor on an i3 for example won't work as it's held by two screws on one side.

Still, I'll do a little review video so we can see what's what.

Well, my order of sorbothane came in on Friday. I got my printer working but it was just super loud. So I printed out some NEMA 23 sized astrosyn style dampers and cut a circle of the sorbothane material to put in between the parts. It has made a big difference at all speeds. I do have another problem though. The vibrations are still travelling through the belt to my X carriage and making the cooling fan really noisy. It seems that the blade and hub part isn't really attached so it bounces around like a speaker. It makes a lot of rattling. Maybe it's just a cheap fan. Anyway, thought you guys might like an update.

Can you post a picture of the machine and one of your motor mounts?

I am still convinced that stepper drivers can make a difference here too. About to order a set of G3D steppers for my CoreXY setup.

---

http://www.marinusdebeer.nl/

Yeah I just don't want to pay the extra cost for the G3D drivers. Did your grounding resistor 4 on the A4988 boards not work out? Or was that for another machine, Ohmarinus? I'll try to post a picture of my machine tonight. I'm still in the process of calibrating and such. I haven't gotten too many good prints yet. I also need to do some cable management. Also, I think I'm having a little Z backlash problem where the first 6-8 layers are squished together and then it becomes normal.

Not the most dazzling pictures of it but it is what it is... A couple of my machine and one of the Astrosyn type mount I managed to print up. I am thinking about maybe trying standalone stepper drivers like the TB6560 stepper drivers. Anyone with experience? Comments on the pictures appreciated.

JTT

How did you stick the sorbothane to the flanges? Do you see any new artifacts in prints?

Those drivers look like they handle more current and are better made than the pololu step sticks, but they're still limited to 16:1 microstepping. I don't think you'll see any difference in performance compared to the A4988 because they're both chopper type drivers with no intelligence.

I use this type [www.sainsmart.com] in my X and Y axes. Sinusoidal currents, resonance damping, microstepping that goes to 256, current to 4A. My Y axis is a special case, but the NEMA-23 on my X axis went from OK with the pololu driver to strangely quiet when it moves once I switched to the DSP driver with 64:1 microstepping.

We just made a simple damper from steel for my Nema 23 0.9 degree 166 oz-in motors, and tested it on one axis.

This made a HUGE difference. It more than tripled the speed of movement at the extreme ends. Note: it's a delta printer - and when the arms near horizontal the motor must move very fast indeed - not normally where I print, but I can print 17" wide that way...

It also eliminated the growly mid-band resonance sound, now the steppers make a pure tone all the way to a very very high pitch. Note that I am using a Smoothieboard so step rates are not an issue!

The feed rate at the extreme outside went from 2350 to 7850 mm/min.

To test this I increased the maximum allowed feed rates in Config so it was not limited.
Without damper: g0 x0 y-214 f2350
With damper on the one axis: g0 x0 y-214 f7850

I've attached a picture. Note that the fan mount was printed long ago when I was having extruder issues...

The basic dimensions are 2" diameter, and near 1/2" thick. We will be testing with a slightly larger diameter (2.4") and I will report if that increases the speed.

I do not think that any fancy vibration absorbing or moving weights are required, just the right mass.

---

My printer: Raptosaur - Large Format Delta - [www.paulwanamaker.wordpress.com]
Can you answer questions about Calibration, Printing issues, Mechanics? Write it up and improve the Wiki!

So this is a flywheel? What is the mass?

Yes, just a flywheel.


We tried the larger one yesterday evening. It did not perform as well, so we made it the same size as the original.
I can't give you the mass yet as I have not measured that yet. I can give the dimensions however:

Smaller/faster flywheel damper:
Diameter: 2" Steel
Flywheel Thickness: .460"
Hub shaft .6" long
hub diameter .550"
hole .152" (to fit the encoder shaft on my steppers)
Without this damper: 2350 mm/min, with this damper: 7300+ mm/min effector speed - moving only one motor, and the tone of the motor is pure.

Larger damper:
Diameter 2.4"
Thickness .507"
With this one the max effector speed was 5650 mm/min max when the delta arms were near horizontal. After we tested this larger one, we turned down it's diameter and thickness to match the first one, as that performance is excellent. A person could turn one down progressively and test it for the optimum for a particular motor, but that takes quite a bit of time (it takes around 2 hours for us to make one and tap it).

Note that the actual speed of the motors is much much higher than this - over 16000 mm/min movement of the delta-arm-cars. I only know this approximately, as I was progressively increasing the allowable axis movement in the Smoothie config, and changed it from incrementally up to 16000, then to 26000 so it would not be throttled. I will adjust these settings back down till they take effect when I've done more testing, and will report on that separately, with video. I still have one more flywheel to make.

OK, you got things moving fast. That's what all the cnc guys posting youtube videos of the dampers did. I don't think it reflects the realistic situation of printing where you have to start, stop, and reverse direction rapidly. What's the print quality like? I suspect that by increasing the moving mass (about 200g for the smaller flywheel) your ability to start and stop the motion has diminished which should show up very clearly in prints. You've increased the moving mass so you'll probably have to run slower (via lower acceleration) than running without the flywheels to maintain the constant print quality.

I think that flywheels and dampers are best applied when the motors are run very fast. When you run steppers at low speeds, as we typically do in 3D printing, especially larger steppers like NEMA-23 or 34 size motors, they're going to vibrate. It would be interesting to gear down a typical 3D printer so the steppers have to run much faster. They'd probably be a lot smoother just because they're going faster.