[Chiuso] Best $7 buck improvement you can make  

Re: Best $7 buck improvement you can make

I borrowed some techniques for vibration isolation from optical labs or DIY STM projects. There are two elements. First one is that for a constant excitation force, the resulting vibration amplitude is inversely proportional to the total mass. So 10 times the mass, 1/10 displacement. For this reason, precision machinery has polymer concrete or cast iron bases. ASMLs latest EUV stepper weighs 100 tons IIRC. So having a heavy stone is a good idea.

Now the second step is adding a decoupling from the stone to the table (mostly it's the other way around, you want to keep vibrations from cars and trains away from your holograpy setup). This is done with a spring - the softer, the better (transmitted force is directly proportional to spring constant). This requirement together with the large mass of the stone above dictates that the spring needs to be much larger than the vibration amplitude.
As spring+mass gives an oscillating system with potentially catastrophic resonance, also a damper is needed to reduce the Q just enough (because a damper causes a force proportional to the relative velocity -> also transmits vibration) so things don't get out of hand in the resonance case.

The cheap and classic solution for this is an inflated bike tube. Air has very low damping (this is why flutes work) and the rubber tubing is adding damping (by tension of the tube) just enough so the setup doesn't wobble too much. What about rubber feet? Rubber in compression mode is surprisingly stiff (Quite some time ago, I did measurements with a shaker/accelerometer to measure the frequency response of such setups) and pretty much like a brick at lower frequencies. Hanging the base from ceiling or a tripod by rubber strings is another classic (and even better), but the bike tube is sufficient for our case. On my desk, it completely eliminated vibration transmitted to the desk (which acted as loudspeaker before).

Only part of the noise is transmitted via the desk. Some is also emitted by the vibrating structure of the printer itself. If the stiffness of the printer from motor to feet would be high enough, bolting the feet down would reduce noise from the non-moving part of the Y and Z motors - but I guess it's probably too soft for this to be effective. The remaining noise is coming from the moving parts.

In my setup, the stone I got was too small for the bike tube (though it fits the MK2S perfectly), so I added an intermediate MDF panel in between (with rubber tubes as isolators - remember, no flat rubber?).

BTW the vibration is caused by the torque ripple of the stepper motors. One part is from the drive waveform - it is usually (also with Prusa) not sinusoidal, but has steps ("mictrostepping"), directly translating into audible noise. Interestingly, even with infinite microstepping (sine drive), most stepper motors will will not rotate with constant angular velocity but still have "steps" in their rotation. This is caused by the structure of the stator and rotor. Some Trinamic drivers (not sure about the 2130) allow to compensate even that - you can download a correction curve which fine tunes microstep positions so the resulting motion is as smooth as possible. As motors have some variation due to manufacturing tolerances, there are limits for this. I evaluated a stepper motor/Trinamic setup for a medical device recently, but went BLDC finally because of noise. I am really curious how the MK3 works out.

- Martin

Re: Best $7 buck improvement you can make

I saw this post and tried out a granite block from work and it helps with the noise as well, main noise I was getting was from the table I had it on vibrating.

Re: Best $7 buck improvement you can make

Well done Martin !

Re: Best $7 buck improvement you can make

I've been looking around for some granite plates, but all I find are the calibration bench version that are in the $400's plus. I'll go see what H.D. has in pavers

Re: Best $7 buck improvement you can make

16 inch stone paver, about 4 bucks
4 big rubber feet, about 3 bucks

Noise, down about 80%
Prints, noticable better

IMG_20170121_175757541_HDR.jpg

Here are someore pictures
https://goo.gl/photos/R3Lts9Qcms4Prz8s6

Jordan, do you still keep the Prusa that close to your computer printer? Any heating problems?

Re: Best $7 buck improvement you can make

That looks mighty cool!
I'm wondering though, how heavy is a typical paver? I'm planning on building a LACK enclosure, and I'm not sure how well the paper honeycomb inside will handle the weight...

I'm adding in with an answer to this. Not how heavy it is but noting that the LACK table will handle the mass completely fine. It can however still put some vibrations into the table as the legs aren't very sturdy so I plan on stiffening the legs somehow.

Anyway, adding the 40x40cm block of concrete really removes all high-frequency mechanical noise from the drivetrain. Incredible how much difference it does, I will completely recommend it!

Re: Best $7 buck improvement you can make

I am not particularly concerned with the noise, I don't think it's too bad.

But we all want better looking prints, so are there any before and after shots?

Re: Best $7 buck improvement you can make

If there is a Granite worktop manufacturer near you, they may have a scrap chunk from where they cut an openning for the sink bowl.

regards, Joan

Re: Best $7 buck improvement you can make

For those in the States, as we are very concrete centric, I have been using the blocks we use to build houses as they are around $1.50 each at our home centers. They also serve as good place to put tools and misc. stuff...

Re: Best $7 buck improvement you can make

Not sure if the pics were for me, but when I asked for before and after shots, I was meaning printed models. 😛

Re: Best $7 buck improvement you can make

when I asked for before and after shots, I was meaning printed models
I did it because of noise only and for that it worked well.
I did not notice any quality difference before/after, but I have not tested this extensively. Probably bed level fine tuning keeps better on a stable basis. No signs of "ringing" with or without base.

Anyway it is not immediately clear to me what effect a heavy base might have on vibration. Imagine two cases:
1 Printer locked down to a 100 ton granite basement
2 Printer floating in space (like Tesla's roadster, just without second stage attached)
Assume we move X +10 mm. Case 1: Only the X carriage moves, while Y is stationary in x direction (idealized). Case 2: The X carriage moves in + direction and the rest of the printer in - direction (reaction force, according to the ratio of X carriage weight to remaining printer weight). This means that in case 2 the x carriage moves less and the rest of the printer more than in case 1. Which case is better depends on the stiffness (and damping) of the structures of x carriage and the rest of the printer. For example, let's say we print some gooey jelly part (I know there is no jelly filament, just making this up), in case 1 the jelly would be stationary and in case 2 it would start to wobble, which is not a good thing. But lets say we have a rather stiff printer (and a non-gooey part), but a long and not well fixed hotend, it would vibrate less in case 2 (as absolute acceleration would be lower).
This is very theoretical though. Fortunately PR got rid of the miserable threaded rods in the MK3, so hopefully the structure is less prone to vibration. The weakest elements will be the drive belts, which fortunately have a lot of internal damping, which helps to reduce vibration. As it seems this is necessary though - with the doubled drive voltage possible acceleration also doubled.

- Martin

Re: Best $7 buck improvement you can make

No problem. I thought I read somewhere that claims of better print quality had been made, so thanks for the clarification.

Re: Best $7 buck improvement you can make

If You drop by your local kitchen countertop fabricators some Friday afternoon about quoting time with a two-four of a local beer, there is a very good chance you will depart with a very nice polished stone sink cutout plug... and depending on their week you might even have a choice of several... I have made a cool coffee table made out of a double kitchen sink cutout of granite and 4 legs epoxied to the corners as well as a bathroom sink cutout of marble for my printer to sit on. These counter scraps are also incredibly flat.

Re: Best $7 buck improvement you can make

Has anyone tried if the MK3 can be made quieter with this method?

Re: Best $7 buck improvement you can make

I used an 18 x 18 inch tile with rubber feet. Same results.

Re: Best $7 buck improvement you can make

I used an 18 x 18 inch tile with rubber feet. Same results.

What do you mean by same results? Same as with the MK2? Or same as without the stone?

Re: Best $7 buck improvement you can make

I notice your trapezoidal nuts are installed the opposite direction as on my MK3. Probably a change between models, but thought I would mention it.

Just an fyi.

Steve

Re: Best $7 buck improvement you can make

I'm using a big paver under my Prusa I3 Mk3 and I put a 12 inch bicycle tube underneath the paver. This makes a very good decoupling, as the resonance frequency is now around 2 Hz for the whole setup. Even in stealth mode, it is much more quiet now!

RE: Best $7 buck improvement you can make

I have a heavy wooden chest of drawers (2 drawers, full of heavy objects) with a sheet of corrugated cardboard on the top. The prusa mk3 sits on there without rubber feet. The wooden chest the prusa sits on has a lip where i wedge in a plank of wood, tight, against the floor. I have carpet.

My thinking is that I want to dampen large movements, and also dampen small movements. i reckon the ideal would be a heavy as solid-surface, with the prusa tied down strongly with stretchy cable. And with a grippy material which prevents it from moving sideways between the prusa and table.

The rationale behind the stretchy cables is that i want to couple the prusa's acoustic vibrations to a surface which will absorb it. i want the opposite side of that absorbing-material to never ever move.

anyway thats how my brain sees the system.