[Arjay]

Folks,

I've been piecing together a design for a cnc gantry router for a few years now. I'm kind of doing it in reverse starting with the z-axis, then to x-axis gantry and finally to the y-axis table. The main reason for this odd order is I have limited space where I live right now, so I'd like to work on something that doesn't take much room, but involves the electronics, motion and the spindle. Working travel will be 1400x2000x275.

In an old post by ger21, he mentioned his Italian made work machine has a 12 foot long fixed ball screw with rotating nut. I gather that the rotating nut design overcomes screw whip.

Out of curiosity, are these ball screw nut drive mechanisms available (at a reasonable hobbyist cost)? If not, has anyone tried making one using the wheel bearing from a front wheel drive car? These bearings are annular and can be had for about 100 dollars US.

[A_Camera]

I can't help out with any ideas about the rotating nut, but there is a long thread on this forum:

Rotating Ballnut - design ideas

It's about rotating nut, so maybe you can find something there.

I understand you start with the Z, and I think that's a good idea. Personally I also started with that one for the same reason you do, it can be built on a desk and not needs a lot of space. There is another reason to start at that end, which is that the Z is the most complicated part and is very critical, also difficult to fix if something needs to be fixed. Good luck.

[Arjay]

I can't help out with any ideas about the rotating nut, but there is a long thread on this forum:

Rotating Ballnut - design ideas

It's about rotating nut, so maybe you can find something there.

I understand you start with the Z, and I think that's a good idea. Personally I also started with that one for the same reason you do, it can be built on a desk and not needs a lot of space. There is another reason to start at that end, which is that the Z is the most complicated part and is very critical, also difficult to fix if something needs to be fixed. Good luck.

That thread is the mother lode of rotating ballscrew nut info. Thanks for the encouragement on my build.

[wizard]

Probably the biggest problem with rotating nut solutions is that you either need to buy a suitable rotating nut and pay for it or you build one yourself. If you build yourself you need access to a decently sized lathe at a minimal. It also likely result in a more complicated design though that depends upon how you implement the machine. Your final machine size is a bit long in the X so it is good that you are thinking about leadscrew whip. Just don't dismiss other possible ways to address whip.

[Arjay]

Probably the biggest problem with rotating nut solutions is that you either need to buy a suitable rotating nut and pay for it or you build one yourself. If you build yourself you need access to a decently sized lathe at a minimal. It also likely result in a more complicated design though that depends upon how you implement the machine. Your final machine size is a bit long in the X so it is good that you are thinking about leadscrew whip. Just don't dismiss other possible ways to address whip.

AFAIU, whip can be mitigated by using two fixed end mounts and limiting rpm. Besides increasing screw dia, are there other ways I'm missing? Btw, what is the typical layout for x and y axis? Is the spindle moving left/right on the gantry considered x?

[louieatienza]

Check out Phife's build thread, I believe he made a DIY rotating nut from off-the-shelf parts from Misumi...

[wizard]

AFAIU, whip can be mitigated by using two fixed end mounts and limiting rpm.

You can limit RPMs by increasing the lead on the screw, that is how far the nut moves per rev. I suspect you already know that.

Besides increasing screw dia, are there other ways I'm missing?

Some have resorted to a sort of idler arraingement midway on the leadscrew that swings out of the way as the nut passes by. I believe somebody on this forum implement such a mechanism.

Btw, what is the typical layout for x and y axis? Is the spindle moving left/right on the gantry considered x?

That is a loaded question and has lead to some discussion here.

I dont buy the left/right movement as a defning way to label the axises on a machine. This mainly due to working on machines with the controls or operator interfaces on wheels or a pendant. How the user sees the axis can vary depending upon the job.

Some people like to define the long axis as X. This has some merit but then we have a lot of routers and engravers with square layouts of the axises. That is the X & Y travel the same distance. So you coild revert to left & right but you still have the issue of whos left and right.

My preference is to start at the top with the Z moving the spindke in and out of the work. This is consistent with most approaches to labeling machine axis. The gantry becomes Y and the base X ( reverse alphabetical order). Doing this we always know that Y refers to the gantry and the X is the base of the machine. Is this perfect or suitable for everybody? Ptobably not but it would make communications a bit easier. In a nut shell saying the Y axis becomes the same thing as saying the gantry. For me it is about making communications easier.

There have been a few postings here from people that don't even respond when you ask them what axis are you talking about. So you are left trying to devine what they are trying to discuss. So in such a case no standard would help.

Since opinions vary here about the only thing i can suggest is to try to remove any ambiguity that might exist with respect to a discussion. If the discussion is about an X gantry, refering to that axis as the gantry should clear things up.

[Arjay]

You can limit RPMs by increasing the lead on the screw, that is how far the nut moves per rev. I suspect you already know that.

Some have resorted to a sort of idler arraingement midway on the leadscrew that swings out of the way as the nut passes by. I believe somebody on this forum implement such a mechanism.


That is a loaded question and has lead to some discussion here.

I dont buy the left/right movement as a defning way to label the axises on a machine. This mainly due to working on machines with the controls or operator interfaces on wheels or a pendant. How the user sees the axis can vary depending upon the job.

Some people like to define the long axis as X. This has some merit but then we have a lot of routers and engravers with square layouts of the axises. That is the X & Y travel the same distance. So you coild revert to left & right but you still have the issue of whos left and right.

My preference is to start at the top with the Z moving the spindke in and out of the work. This is consistent with most approaches to labeling machine axis. The gantry becomes Y and the base X ( reverse alphabetical order). Doing this we always know that Y refers to the gantry and the X is the base of the machine. Is this perfect or suitable for everybody? Ptobably not but it would make communications a bit easier. In a nut shell saying the Y axis becomes the same thing as saying the gantry. For me it is about making communications easier.

There have been a few postings here from people that don't even respond when you ask them what axis are you talking about. So you are left trying to devine what they are trying to discuss. So in such a case no standard would help.

Since opinions vary here about the only thing i can suggest is to try to remove any ambiguity that might exist with respect to a discussion. If the discussion is about an X gantry, refering to that axis as the gantry should clear things up.

Okay, assuming x as forward & back and a y gantry, with the operator standing centered in front of the x axis, what corner typically is home? Far left? Close left? Something else?

[Arjay]

Check out Phife's build thread, I believe he made a DIY rotating nut from off-the-shelf parts from Misumi...

Would you mine posting a link to that thread? I tried searching and wading through the results but I couldn't find a build thread started by phife.

[louieatienza]

Would you mine posting a link to that thread? I tried searching and wading through the results but I couldn't find a build thread started by phife.

Search user names, then click on the name go to statistics and click find threads started by phife

[ger21]

Okay, assuming x as forward & back and a y gantry, with the operator standing centered in front of the x axis, what corner typically is home? Far left? Close left? Something else?

Wherever you want it to be.
I don't follow where you are standing. I like to stand so that X is left to right, and I make my origin in the left corner closest to me.

[turbocad6]

it is up to you to decide where you want the home position and what quadrant you want to be working in. personally I like working in quadrant 1 meaning all of my X and Y movement goes in the positive direction fromX0Y0 home which would be in the lower left corner, but different machines can be set up in different ways and for many that work with a vise most times this can be an issue because the fixed jaw is not on the "bottom" it is usually on the top so many machines are set up with top left or top right as X0Y0. many times this can be dependent on how your vise is mounted and where your fixed jaw faces.

on a milling machine most times the table is wider left to right when you are standing front of it, like a landscape orientation, so the long axis would tend to be the X but on a router table this is usually opposite, meaning they usually tend to be more set up in a portrait orientation, meaning the longer axis tends to be more the axis that goes towards and away from you more than left and right when standing in front of the machine so there the longer axis can then be the Y instead of the X.

this actually comes down to perspective and what you perceive as the "front" of the machine. in the end this comes down to perspective and view. for me personally in order to be able to work in quadrant 1, yet still have my gantry back away from me when homing I actually have my machine setup in such a way that my work is all machined upside down, meaning the bottom of my work piece is actually towards the top of my table. the "front" of my machine is actually at the back of my machine.

I know this can be a bit confusing and I have had several friends that I have tried explaining this stuff to and it can be hard to explain with just words because again, perspective comes into play so the easiest way I have found to explain it is by just showing you the cartesian co ordinate system diagram, which is what this is all based on. in reality it is kind of simple once you look at the diagram and realize that the center of the + is home, and depending on which corner you choose that to be will determine what quadrant you are actually working in.

*if you home at top left then you will be working in quadrant 4, all of your moves will be in the X positive but Y negative

*if you home at top right then you will be working in quadrant 3 and all of your moves will be in the X negative and Y negative region

*if you home at the bottom right then you are working in quadrant 2, al of your moves will be in the X negative Y positive

*and if you home at bottom left then this is quadrant 1 which would make all of your work be in the X positive Y positive region.

there is NO right or wrong way, it comes down to perspective and understanding and following a few simple rules, such as going from left to right is always a positive move and also going from bottom to top is always a positive move even if you happen to be working in a negative region, then "positive" would simply mean "less negative" UNLESS like me, you chose to work upside down which may sound a bit odd at first but it has turned out to be the most intuitive way for me and my own perspective. also of course universally it is understood that going up is almost always a Z positive move so that is usually much more universally understood...

the problem with communicating with others sometimes is that there are 4 different and unique quadrants that one could be working in, so just saying X positive or Y negative can mean different things to different folks depending on there perspective in relation to how there own machine is set up.

look at this diagram for a bit and everything should really start to fall into place and sink home with understanding this. after typing out this post I actually thought to make this info as it's own thread because honestly this is something that trips many people up but in looking around the forum I honestly don't even know which section of the forum this post would even be most appropriate in

[louieatienza]

Your work and home coordinates need not be the same. Home can be in a convenient location for you to do tool changes, use a tool length setter, or allow access to the work surface.

The convention is, X is left to right, Y is toward and away, and Z is down and up, and those are the typical directions from negative to positive. REGARDLESS OF HOW YOUR MACHINE IS CONFIGURED. This way, the position you present work to the machine is exactly the same as what you have in CAD and CAM. The machine DOES NOT CARE whether your gantry is X or Y, some machines (like CR Onsrud heavy duty) have a long gantry, and the material is fed through the front. So the gantry is X. A Biesse has a shorter cantilevered gantry attached from the back, and I think Morbidelli is similar. The "front" would be the X and the gantry would be the Y. You could potentially reverse the axes, but then your computer monitor is not configured that way, and it would be odd to do so.

[Arjay]

Search user names, then click on the name go to statistics and click find threads started by phife

Found it from a link in another one of your posts. http://www.cnczone.com/forums/diy-cnc-router-table-machines/147509-phifes-fixed-gantry-ridgid-z-rotating-nut.html

[Arjay]

After reading the other threads about the rotating ballscrew nut and giving this some thought, I have a couple of questions with regard to oil seals.

From looking at bearing reference catalogs, higher bearing speeds are obtained when using oil rather than grease. My servo motors top out at 5K rpm, so I'll need to use oil to reach that speed for the bearing I'm planning on using.

The shaft that attaches to the ballscrew nut is 304 stainless steel. It holds a double wide angular bearing as well as a needle bearing. In addition, this part has a 80mm surface where an 80x100x10 oil seal would ride on.

Now finally the questions:

Will I have wear issues with the lip of the seal cutting a groove into the stainless? If this is an issue, would hard chroming that part of the shaft be a good solution?