[powerfade]

Hey there all.. I have a problem I have fighting for awhile and was wondering if anyone might have any ideas. I built a 5'x10' CNC router about 4 years ago that ran on 1" ball screws. It worked pretty well, but I wanted a better cut on acrylic. I have recently switched over both axis to rack and pinion. The y axis is great, acrylic edges are super smooth. The x axis however, did not show a whole lot of improvement. The specs on the machine are:

* gantry probably weighs about 200 pounds.
* Since I had the ball screw running down the center, I took all of that out and put a bar down the center of the table and am running the rack and pinion down the center.
* I have the older round thk rails and bearings.
* I would imagine the whole table with everything is around 1000 lbs.
* rutex drives driving pretty big servo motors
* the machine is primarily bolted together

I have attached some pictures for you to look at.

Any ideas on how I might eliminate the vibration on the x??

[CarveOne]

Some things to check for:

1. Is the pinion gear exactly 90 degrees with respect to the rack?

2. Is the pinion bottoming out in the rack grooves or vice versa?

3. Are you getting harmonic vibrations in the gantry cross member under the table because it is not stiff enough for this type of drive system?

CarveOne

[powerfade]

I took a look at the teeth and they are meshed together well. From what I can tell the the rack is at a 90 degree from the pinion. As far as the cross member, I am not sure.... When I hop under the table and move the gantry back and forth, I can feel vibration in the 1/8"x2"x2" bar that the rack is mounted too. I may need to support the 1/8"x2"x2" better.. It's supported at both ends of the table with a bolt on each side and nothing in between, almost floating if that makes sense.

[CarveOne]

I took a look at the teeth and they are meshed together well. From what I can tell the the rack is at a 90 degree from the pinion. As far as the cross member, I am not sure.... When I hop under the table and move the gantry back and forth, I can feel vibration in the 1/8"x2"x2" bar that the rack is mounted too. I may need to support the 1/8"x2"x2" better.. It's supported at both ends of the table with a bolt on each side and nothing in between, almost floating if that makes sense.

It could be the gantry cross-member and/or the tube. The rack is in the most susceptible area for flexing of these parts. See if you can figure out a way to dampen them by temporarily wedging something against them (so that no calamity happens) and try it again.

If the pinion is not square with the rack the teeth will chatter. If the gear engagement tight it may also cause chatter if either side has teeth that are hitting the very bottom of the grooves. I have a rack that had teeth that were too tall because the tops of the teeth had rolled "wire edges" on them. I just cleaned them down with a whet stone so that there is a small gap seen between the teeth and the groove. This was a 1/2" rack, and the rack ended up measuring 0.500" when finished with the stoning.

CarveOne

[Mariss Freimanis]

What is your gearing reduction ratio? Your X axis motor is shoving around a much greater mass than the Y axis motor. That means the X axis has a greater inertial load which can lead to servo instability if the gearing isn't correct. Trouble starts when the load moment of inertia is more than 10 times the motor's moment of inertia. Increasing the reduction ratio decreases the inertial mismatch between motor and load. Ideally the motor should be turning at its rated RPM (80% of no-load RPM) when your mechanism is moving at its maximum design speed.

Mariss

[powerfade]

Thanks for input everyone..

CarveOne: I am going to try and support the rack as you mentioned by temporarily wedging something in there. The gantry cross member is 1/2" aluminum plate with a 1 1/4" solid square aluminum across the top for extra support.

Marriss: I have a 7:1 reduction on the motor. They are fairly large motors. If I am not mistaken I think about 900oz in. I should mention that vibration is not excessive or noisy. It is actually pretty nominal for the most part, but definitely noticeable. 1/4" acrylic with a 1/4" bit leaves a perfect edge on the y axis and a small but noticeable waivey cut on the x axis.

Actually, in addition to the waivey x axis cut, when I change directions from x axis to y axis or visa versa on a 90 degree cut it vibrates quite a bit (spindle shakes from the change in direction) and leaves a pretty noticeable waivey cut.

Thaks for all of the input...
Chris

[CarveOne]

Thanks for input everyone..

CarveOne: I am going to try and support the rack as you mentioned by temporarily wedging something in there. The gantry cross member is 1/2" aluminum plate with a 1 1/4" solid square aluminum across the top for extra support.

Marriss: I have a 7:1 reduction on the motor. They are fairly large motors. If I am not mistaken I think about 900oz in. I should mention that vibration is not excessive or noisy. It is actually pretty nominal for the most part, but definitely noticeable. 1/4" acrylic with a 1/4" bit leaves a perfect edge on the y axis and a small but noticeable waivey cut on the x axis.

Actually, in addition to the waivey x axis cut, when I change directions from x axis to y axis or visa versa on a 90 degree cut it vibrates quite a bit (spindle shakes from the change in direction) and leaves a pretty noticeable waivey cut.

Thaks for all of the input...
Chris

The mass of the gantry was why I was focusing on the mechanical side of things that might be flexing and potentially being misaligned a little. Mariss covered the electrical side of it that I am far less knowledgeable about.

Hopefully one or the other will lead you to a workable solution.

CarveOne

[Andy990]

Hello! You could try placing pieces of dense solid rubber mat under the machine where the legs go to the floor. You could also try taking a bicycle inner tube partially filled with water in various accessable spots on the machine. If you have any unused holes in the steel frame anywhere attach a piece of rubber to try and absorb the vibration, or clamp on a piece of rubber temporarily. Could take a tire filled with water and suspend it underneath the machine using rope. Could clamp on a piece of rubber on the pillow block and see if that works.
If you can measure the frequency of the oscillation..that may provide a clue to the solution if you can not find it otherwise. It may be sinisoidal or non sinousoidal.
Is there any way to attach a circular piece of rubber to the outside of the gear on the rack and pinion? Take a piece if fishing line or string with a leasd sinker and suspend it at various points on the table to locvalize the problem. Take a piece big of lead and sit it somewhere.
You could rig up a spring mass damper and try and attach that also. Could use a very thin dense piece of rubber between the servos and the servo mount.
Add a piece of rubber bracing to the table. Take a spring loaded clothes rod and wedge it under the table with rubber on both ends. Attach a tuning fork to the machine to convert the vibration to sound. Could also try dry friction damping and just set a heavy piece of rubber somewhere. Find some big rubber bands and try placing them (lots of them) at various spots and acrooss various position and connectong various parts with them..might help isolate the problem. Or a car shock absorber attached somewhere to somewhere.Take a big wad of rubber bands and put them around the servo motors. Or a storm door opener closer attached somewhere to somewhere. 3M makes vibration damping tape. Temporarilly tape some rubber to the end of the x axis ballsrew.put some paper around the ballscrew to protect it first. to the part of x axis where you never go to. Once you locate it you can go for a more permanent solution. Take a piece of pipe and use it as a sounding tube to listen to pillow block bearings and other bearings,,or a mechanics stethoscope

Just some thoughts
andy

[Andy990]

Hello! Here is a cheap way to possibly quantify the vibration for numerical analysis. and load the waveform into a computer go to:http://www.hometheatershack.com/roomeq/ and download the free room eq sound analysis software..hook up a microphone..any cheap..cell phone ect.. mic will work! to your computer..let it analyze the sound and display a numerical and graphical representation of the sound.
Its also good to figure out how to quiet down a room with a bunch of machinery running. Three pieces of different material deaden sound best..rubber is usually one and wood is another..wouldnt be too hard to surround just the router so you dont have to listen to it scream all day or at least mitigate it......that is of course if it makes noise..
You can also use this to analyze bearings and servo motors as failure analysis or confirm correct operation with a stethoscope ....just a thought
andy

[eloid]

is you pinion sprig load the rack sounds like you sqaure on the rack and it needs adjustent and to be spring loaded to the rack?

[Andy990]

Here is a link to a free interactive vibration problem solver for machinery. You click on various word descriptions of the suspected problem and it shows you what to do
http://www.vibanalysis.co.uk/

good luck
andy

Quote: Dreese Lab becons those with intellectual acumen and rewards them

[turmite]

Hi powerfade,

I sent you a pm but promised I would post here as well as it might help others.

I think part of both your original and current problem is the amount of over hang you have on your z axis. Do you have a way to measure that from the center of your y , or x, depending on which you have your z mounted? I have a 5X10 r/p machine, and I have vibration as well, but my machine is much more massive than yours.....so I don't really think it is the r/p as much as the overhang.....and I have that too.

Mike

[powerfade]

Thanks for all of the input everyone...I will try them all trust me... I'll let you know the results and if anyone else has any brainstorms I am all ears.

Answer to question 1: I do not have the gear spring loaded into the rack.... yet anyways?? I think I will have to do this though... It seems to be an absolute..

Answer to question 2: I would imagine the center of my spindle is probably a good 12" - 14" from the center of the gantry. How bad is the vibration on your machine??? Does it affect the cut significantly??

Thanks again...

Chris

[CarveOne]

Thanks for all of the input everyone...I will try them all trust me... I'll let you know the results and if anyone else has any brainstorms I am all ears.

Answer to question 1: I do not have the gear spring loaded into the rack.... yet anyways?? I think I will have to do this though... It seems to be an absolute..

Answer to question 2: I would imagine the center of my spindle is probably a good 12" - 14" from the center of the gantry. How bad is the vibration on your machine??? Does it affect the cut significantly??

Thanks again...

Chris

You definitely need the spring loading, as this will help tie the rack and the pinion together. Make it adjustable in tension with a turnbuckle. I just assumed that you already had one on it.

CarveOne

[Mike Nash]

I'm wondering if you are getting racking when moving the gantry as you implied it did the same thing with the ballscrew. Check this thread http://www.cnczone.com/forums/showthread.php?t=51485

[powerfade]

Merry Christmas and Happy Holidays to everyone... So I think the first thing I am going to need to do is definitely spring load the pinion to the rack.. How strong a spring would you say would be sufficient?? I'll do a search a bit later to check on the site to see what other people have used, but if anyone has a suggestion in the meantime. Something like a valve spring or something a bit lighter??

Mike I took a brief look at that link you sent. I have been concerned about racking and that looks like a very clever solution..

Thanks everyone and have a great day off....

[CarveOne]

Merry Christmas and Happy Holidays to everyone... So I think the first thing I am going to need to do is definitely spring load the pinion to the rack.. How strong a spring would you say would be sufficient?? I'll do a search a bit later to check on the site to see what other people have used, but if anyone has a suggestion in the meantime. Something like a valve spring or something a bit lighter??

Mike I took a brief look at that link you sent. I have been concerned about racking and that looks like a very clever solution..

Thanks everyone and have a great day off....

Any kind of very stiff spring with a turnbuckle in series with it will work for a test to see if it works, but my cncrouterparts drives use a spring that with 0.600" long, 0.750" diameter, 6 turn, coil of 0.105" diameter wire body and a turnbuckle that has a 1.750" long body. There is a photo in my build log in post #434, middle picture, that shows it. Fastenal.com is a place you can look up an equivalent spring.

CarveOne

[CarveOne]

FWIW, I was at the local Lowes this evening and in the coil spring display of the hardware section is a pak of two springs with the same dimensions as my previous post. Same wire diameter also. Price was a little less than $3.50.

CarveOne

[jsheerin]

After you spring load your pinion, if the problem doesn't go away I'd look at racking of the gantry and then deflection of the rack. Most of the rack setups I've seen have the rack solidly mounted to a fairly massive surface. Yours looks like it can bounce all over the place. Think about how much force is going into accelerating and decelerating the gantry. Then realize that that force will try to deflect the pinion out of the rack (using the inclined planes of the rack and pinion teeth). So spring loading would help reduce the movement of the pinion relative to the rack, but if the rack can move away from the pinion, that's not going to help things.