The Meticulous Z-Axis

It doesn’t matter terribly to me. I never changed it from the default. I designed the new sled for the community, so giving it public domain dedication is fine for me. I’d rather not have someone else making money off of my idea, though. I don’t know too much about the legal aspects of each of the different dedications.

We should probably add your stl files to the community garden page, though, so people can also get them there.

What an awesome design! In the 3d model the ring is mounted at the maximum extent of the L brackets. The standard z axis L brackets were not designed for lateral forces and look somewhat flimsy. From an accuracy point of view, minimizing flex and maximizing rigidity should be the goal. Using beefier L brackets, and/or mounting the ring closer to the sled would achieve the goal. Hence the question about minimum ring to sled stand off distance and opinion about L bracket rigidity.

1 Like

I recently bit the bullet and ordered parts to create my version of the meticulous z axis. I wanted to be able to order all my parts from Amazon so I made some modifications to the list. I ordered a 10mm wide belt and 16T pulleys to match. Here are the links FYI.

16-Tooth Timing Pulley for Stepper Motor

Loop Belt Timing Pulley Tools 10mm Wide

Lead Screw, Bearing, Rod, and etc. kit.

11 Likes

I have recently completed the upgrade of my Z axis using mostly the @MeticulousMaynard design. I still need to find the right weights and set balance correctly, but it moves wonderfully.

The open viewport is fantastic and tool bit changes are so much easier. I had problems with the wood clamps so I added hose clamps to the outer rim and this solved the problem.

Thought I would share a few pictures.










The bottom HDPE slide just before installing.

13 Likes

Just wanted to thank this group for all the work. I was able to build one almost exactly per the plans at the time I started. I did run into the problem mentioned earlier of the wooden clamps interfering with the ring system. My solution is similar to one given, although I put 2 5" 5/16" inch bolts through both clamps. I felt this would be more stable

4 Likes

Here is a closeup of the clamp solution.

5 Likes

Looking good! I always like seeing the solution that people come up with for the issues they run into.

Not sure you’re going to want to make new clamps, but I had modified the spindle clamp design to make it work better with the ring system. They’re the ones circled in green below:

image

It looks like you’re using high-quality plywood, but it still may also be wise to machine the new ones from solid wood rather than plywood. There’s been some discussion in this thread about the voids in plywood reducing the strength of the clamps, which could result in them failing. This would be even more an issue with the new design, which is thinner than the old one.

Happy chip-making!

7 Likes

Thanks for the input. I was going to go with the new clamp design, but saw someone else had some trouble with that as well. Probably from using cheaper plywood. As you noted, we are using very high grade plywood, and there are no voids in it. I considered 3-D printing the new design, but my printer wasn’t big enough. We just got a new printer, so I will go that route if this fails. For now it feels very secure.

3 Likes

I think I already know the answer but would appreciate outside input. Is it required to install the dust chute prior to assembley of the wood parts?

Yes, with the design of the dust chute, it is required to install that to the base before attaching the spine for the Z-axis.

I was just looking through my instructions in the Community Garden for assembly and realized they’re terribly out of date. I will need to fix that, hopefully sooner rather than later. I’ve never expressed the order of operations here, so I can see where it would be confusing.

2 Likes

I’m sorry, my post was not meant as a complaint. Just prioritizing, I will need to set my new 3D printer up before I finish the Meticulous Z axis is all. Personally I am grateful for the work you have put in and shared!

2 Likes

I didn’t see it as complaining at all. I was thinking as I was writing my reply that I had made note of the order of operations when I wrote up the instructions. So when I went to look at it I realized that they were written for the old configuration and had no mention of the order. Not too much has changed in the assembly process, but it still needs to be revised.

3 Likes

I wonder if this would fit within the ring and be able to sync up with Ground Control.
https://www.mlcswoodworking.com/shopsite_sc/store/html/smarthtml/pages/powerlift_pro.html
???

I have seen 8mm rods deflect pretty easily. With the weight and torque of the router, is anyone seeing that with this z axis?

1 Like

I’ve had it with z-axis troubles.

I’m hoping that a meticulous z-axis setup will improve reliability. A while ago I bought this beast on e-bay:


So, now I’m going to take the plunge (that’s a z-axis joke), and build it out.

I’m sure I’ll be back in this thread with a lot of questions. Wish me luck!

5 Likes

i see there is a design with “new clamps” is there a new file uploaded outside the community garden? the file that i downloaded does not show these clamps in it.

also i downloaded the large master file from the community garden however when i upload the .nc file into ground control it does not import to scale. i cut the entire file out but it made everything much larger than it should be. closer to a 22" sled than an 18"… unless this is correct and i didn’t read it. there are 260 posts in this thread so forgive me if i missed it.

1 Like

Sorry guys, missed a couple of comments here, will do my best to answer them now.

I’m not sure, but my theory was that the Maslow has low side-load for a CNC router given the typical 1/4" step-down and the low feed rate. I do plan on putting this design through its paces to find out if the guides will deflect from the side-load, but I need time to get back to working on my machine. I have a couple 1/2" bits I’d like to try and do full-depth cuts, so I’ll really get a chance to stress test the design. If the rods are in fact too flimsy for our applications, I plan on upgrading to these:

https://www.aliexpress.com/item/4pcs-100-original-HIWIN-linear-block-MGN7C-MGN9C-MGN12C-MGN15C-for-mini-linear-guide-for-CNC/32918140567.html?spm=2114.search0104.3.114.15a314ccOzeSIb&ws_ab_test=searchweb0_0,searchweb201602_3_10065_10068_10130_10890_5730315_10547_319_10546_317_10548_10545_10696_453_10084_454_10083_10618_5729215_10307_538_537_536_10059_10884_10887_100031_321_322_10103,searchweb201603_2,ppcSwitch_0&algo_expid=cf3b7b0b-fb3b-4e1f-a63c-fe6b4a531cd4-16&algo_pvid=cf3b7b0b-fb3b-4e1f-a63c-fe6b4a531cd4

They’re more like what a gantry-style machine would use, although they up the cost. I was trying to make this as cost-effective as possible.

I feel ya, relying on the Rigid router base can be incredibly frustrating.

I’d love to see that massive axis slide set up! That would certainly solve the above deflection worries. :wink:

The one thing I’d want to look at is the location of the spindle on that. If you cut the DXF as is, the spindle could be either to high or to low, in the Y direction on the machine. That could result in a significant loss in accuracy. In the current design, the spindle centerline is 118.11mm to the face of the spine, where the slide would mount to. If the difference isn’t too much, you can probably compensate for it with the spindle clamps. If you’re familiar with Fusion, you can download the model I made, and add in your axis to see what changes you will need to make to get it to the right spot.

The DXF file with these clamps is in the community garden folder. They are in 2D CAD Files/Prototype Sled Nested V2.dxf. See the below image for location inside the CAD:

That sounds like it could be a scaling issue, but it’s really hard to say without being at your machine. The sled in this design is 500mm (19.685 in) to make room for the gussets. What you got is 112 % larger than the correct size, which means that the plywood won’t fit in the slots correctly. I can’t remember if ground Control can scale G-code or not. Couple of ways (I think) this could happen. First would be calibration, where it would be distorted in the X and Y dimensions. The other would be using a different tool than I programmed it with, which is a 1/4" bit. The latter I think would be less likely, because you wouldn’t be that far off.

1 Like

Thanks for the tip, @MeticulousMaynard. I was planning to do just that with the y-axis. The rail is 1.66" back to front, so I’ll compensate by making the clamps a little thicker.

1 Like

Thanks for the reply. The side load may not be that high, however, in my mind it would be magnified but the weight of the gantry and router. Maybe I am thinking about this incorrectly. The MGN linear rails are nice but I have had some hit or miss experience with them. The cheaper ones seem to “stick” in place. By that, I mean it takes a bit of force to get them moving but once they do then they go smooth. For 3d printers, this initial jerking motion of getting the rail slide moving caused some bad layer alignment issues. I don’t think you would see that problem here. Any linear guide (like the MGN or the rod type linear rails) is at the mercy of the flatness of the wood. In the case of the Meticulous Z axis design, if the wood isn’t flat then the rail that is bolted to it will not be straight.

I decided to go with a C-Beam setup. I have seen some other CNC machines use C-Beams as well. Waiting for it to come in the mail. I planned on 3d printing the mount for the beam along with making alignment pins that would precisely align the mount to the sled. I was also going to 3d print the router mounts as well. The Maslow is not a high precision machine but I figure anything I can do to help it would be good.

1 Like

According to the online Fusion 360 model, the distance from the front of the z-axis spline to the front of the z-axis carriage is 2 inches. As long as that matches the DXF file, I’m good. Since my rail is 1.66 inches thick, I’ll need to make the clamps 2.00-1.66 = 0.34 inches longer (assuming the clamps are right against the z-axis carriage on my rail).