The Meticulous Z-Axis

has anyone tried making a modified Meticulous Z axis with 200mm rails instead of the 300mm rails? The main issues seem to be
#1. linkage or ring kits that are about 4" tall due to high center of gravity mean the gantry plate needs to be extra tall/long so router can be placed at very bottom, thus increasing height
#2. large 3 to 4" tall router clamps mean that router is again pushed upward

Using a steel sled would reduce ring/linkage height to about 2" and using a metal clamp about 2" tall would allow lower placement of router/spindle.


I think I will try but use the above anti backlash nut seems easier to install as well.

I like the concept, but it seems too tall. Any other design considerations I am missing?

The point you bring up about the axis height is a valid one. It seems like that seems to be the biggest design flaw that I’ve seen come up from supporting my end users. I don’t think it’s been show-stopping except in one issue I’ve seen so far, but it’s definitely on my short list of improvements.

To be honest, I wanted to make sure there was enough clearance above the sled top so that the end user could access the router without taking off the machine. There currently is 57mm (2.244in) of clearance between the collet nut and the dust collector window. Especially if a long tool is being used, that’s already not a lot of space to make a tool change. This issue gets complicated by both the ring and the 45* linkage system.

That being said, if we redesigned the spindle clamp so that the router is easily removable, this issue of clearance may be a moot point. I think it’s possible using a 3D printed snap-lock system, similar to the one on the original router base. Of course, that makes the design rely even more heavily on 3D printing.

I have also considered a metal sled as an option to lower the CG of the assembly. I’ve been working on an all-aluminum version off and on, and that’s one of the design considerations. It’s not as easy to get the same weight as the steel one, but I think I have enough aluminum laying around that I could stack a layer or two to bring up the weight.

The anti-backlash nut is something that I’ve looked into. The one I was looking at was a bit different, but it has a similar result. The backlash I’ve measured in the cheaper lead screw nut is pretty minimal, though. It got as much as 0.04mm, which I don’t think is significant enough in the majority of use cases for the machine. If a higher precision is needed than that, though, it’s a easy upgrade to make.

In short, much of the design considerations were made to make it as easy as possible for an end user to make an improved system using the stock Maslow and other tools that makers have common access to. So many of my material choices were made around what materials they would have easy, relatively cheap access to. All aluminum construction would get costly very quickly, both on material costs and time costs. Steel is simply out the realm of what I would consider a workable material for most. Sure, I could plasma cut something but not everyone has easy access to those tools.

Now, if I were able to make some version of an actual product for this, that opens a lot more options. It’s something that I’ve been working on a plan for but haven’t had the time to pursue more.

1 Like

Logical and fair answers. Thanks for response

2 Likes

I’m designing such a version at the moment: https://a360.co/2UaC3tp

Still rough around the corner: No dogbones, no mounting holes for linear rails and spindle adapter, wrong pulleys… But ~100mm shorter than the original one. :slight_smile:

3 Likes

looks nice. with smaller routers a shorter axis is easier. the weight and bulk of the popular ridged r22002 makes it more challenging.

2 Likes

The r22002 is far more powerful than the maslow really needs, a good 1HP trim
router is a good size for the bits and feed rates we commonly use (don’t get the
1/4 HP harbor freight cheepie), a good one will be ~$100, so not a lot cheaper
than the ridged, but smaller and lighter.

David Lang

1 Like

As long as the RPM can get down to at least 10k, yes, that would be a good choice.

Edit: In my option, though, I think a small, air-cooled spindle somewhere between 0.75-1.5 kw is the best choice. Most come with a more elegant way of mounting the spindle to the carriage, and they’re designed to hold up better to the long run times of the Maslow

While I agree palm routers should be used, the lifetime warranty for the ridged is very nice if you remember to register it online. Perfect for those that put a lot of hours on their machines.

Without the reserved space for the linkage system its easy to save some Z height. :slight_smile:

good point, I didn’t realize some of them have a min speed of 16k rpm

https://www.rockler.com/makita-rt0701cx7-1-1-4-hp-compact-router-kit would be a
good choice (not significantly cheaper than the ridgid, but smaller and lighter)

not that lighter means much when we are adding bricks to make the sled heavier,
but it does make the weight lower on the sled.

David Lang

I ran into that problem when I was helping someone out with a build awhile ago. I didn’t realize it myself until I looked at the specs for Rigid’s palm router. Suffice to say, I was a little annoyed they were so high speed. I just had a thought though, I wonder if they also have adjustment pots on the variable speed board like the R2200 has. That could solve that problem fairly easily. :thinking:

Most certainly. The big takeaway is that this would solve is that it lowers the CG, which we desperately need. The other potentially useful benefit would be that we may be able to push the Z-axis motor faster, which would speed up Z travel times. I do think that the current release of the Meticulous Z axis does a good job with this, but it would be awesome to get it going even faster.

For the sake of sharing information: Made a little more progress on the fitting of my Z-azis. I really like the printed clamp but it looks like (according to my measurements) the bearing on the ring system won’t clear the latch on the clamp by about 1/8th of an inch. Real bummer.

3 Likes

a belt sander or grinder would trim off 1/8 easily

2 Likes

Yeah that is what I have been considering doing, it will probably need to be 3/16 removal though, 1/8 is pretty much at zero. At some point I may modify the design of the clamp, if I do I will share with the community.

too bad you can’t rotate that bolt part of the clamp to the front/top where the ring is bolted so the rollers would never touch it.

1 Like

Another update:

So I trimmed ~3/16 from the clamp and solved that issue. I opted not to sand or grind it off as this is printed PLA, (layers of plastic). I used the bandsaw and sealed the cut area with CA glue.

I was then having trouble with the new Z-axis motor, nothing responded as expected even with the correct values entered. So a little digging and I came across this post:

I was having similar symptoms and my motor was manufactured by the same company and included the same diagram.

I can comfirm the correct wiring for this motor (in addition to the one in the post above) is same color to same color with the exception of the orange (original maslow z-axis cable) to Blue (motor wires).

So at this point things are getting better but the z-axis is not accurately traveling the distance I ask it to. I tried pretty much every work around people have talked about on this forum. It was moving about twice the distance I comanded.

I ended up doubling the encoder steps (after trying everything else in vain) and was able to produce repeatable accurate results and a speed more along the lines of what I expected.
The documentation: (Metal DC Geared Motor w/Encoder CQGB37Y001 - CQRobot-Wiki)

It calls for 2688 steps per revolution. I double it to 5,376. Then I was getting 18 mm when I asked for 20mm. So 18/20=0.9, 1-.9=.1, 1.1x5,376=5,913.6, I rounded to 5,914. I know some of y’all are really sharp on this stuff, any suggestions are greatly appreciated. 5,914 is working really well and produces accurate results. The pitch on the lead screw is 8mm per revolution. Both Pulleys are 16 tooth, 8mm bore.

Here is the equipment I used in case it helps someone else.

https://www.amazon.com/gp/product/B01B5QT71K/ref=ppx_yo_dt_b_asin_title_o06_s00?ie=UTF8&psc=1

https://www.amazon.com/gp/product/B074J7BBX8/ref=ppx_yo_dt_b_asin_title_o02_s00?ie=UTF8&psc=1
(I only needed the female connector to solder the motor leads to in order to plug in the original z-axis cable.

I ordered all this stuff back in 2018, wow, I stay pretty busy but I didn’t think is was than long ago. But, except for changing some of the calibration numbers, (I still need to run calibration, but may just need to double check the distance from chain to bit center) this z-axis project is done. This setup is faster and more accurate than the stock z-axis but, not lighting fast. This is what I was hoping for and I am pretty satisfied so far, been a long road but I think I am there.

The clamp I printed and the discussion can be found earlier in this thread, if I get around to modifying the clamp I will post it on this thread.

Cheers!

3 Likes

Need to say thanks to @MeticulousMaynard, I am really satisfied with the Z-Axis. It operates just like I wanted; smooth, precise, accurate, and plenty fast. I am beyond happy I finally found the time to get this project done. I am looking forward to getting a lot more use out of the maslow now!

First project with the new Z-Axis:

4 Likes

This is with using 200mm rails. The 3/4x5.5"x 7.5" board the router attaches to takes up a lot of vertical space so the z axis adjustment range is only about 1.5". I think the next one I try I will make is shorter, maybe 4.5x7.5 wide. Just used some scrap 2x4’ s cut at an angel to mount it to my sled.

3 Likes

What is your plan for bit changes? (no snark intended)