Inventables Community Forum

JBusby's Linear Z Axis Upgrade

So awhile back I purchased a linear Z axis from microcarve on eBay before they jumped their prices up. It’s the same unit that Phil and others are using, I just haven’t had the time to do the upgrade until now.

Here’s my adapter plates that I designed in Vcarve:

The top plate will connect to the Z axis and I have a CNC Router Parts DW611 mount (here that will mount on that plate.

The bottom plate is designed to use the 4 M5 holes in the Xcarriage to secure the plate flush to the Xcarriage (M5 thru-bolts and lock nuts). It has holes (will be threaded) for T-track to mount on the plate for the SuckIt dust boot. There are 6 1/4"-20 holes on the plate that will be threaded and the linear Z axis will screw there.

I started cutting the aluminum tonight.

Feed: 48 ipm
Plunge: 15 ipm
DOC: 0.006" for the first carriage plate holes, 0.012" for the spindle mount plate.
Bit: 1/8" upcut 45 deg helix ZRN coated from Kodiak (here)

Here’s where I left it for tonight:

Not sure if it’s the bit or what but wow, it was just cutting this aluminum like butter. Even with what I thought was aggressive settings, it still did just fine. The cut-out toolpaths will be a bit more taxing than the holes. Using the climb cutting option in Vcarve. I did some measurements using my caliper to compare to the design and it’s almost dead on accurate. I ran into some weird UGCS problems so I called it a night. Not sure if it’s UGCS or the Xcontroller or the RaspberryPi. Also, not too fond of how Vcarve retracts to the safety height after each pass, When pocketing those holes, it was a lot of movement that other toolpaths may have eliminated. It still was pretty quick even with the lower DOC, as each hole was taking about 4 mins per hole @ 0.006" DOC. For me, it’d probably take that much time for me to align and drill on a drill press. At the deeper DOC, it took like 20 mins for the top counter-bores and 10 mins for the thru-holes at the top.

I don’t have a clamp at the 0,0 (using some carpet tape on the backside to hold it down towards the middle) and the material is off the board more so my Z zero is higher which has led to a few holes not getting cut all the way through. I’ll make some cleanup toolpaths to run and it should be good.

I still need to cut the 4 spindle mount holes then the profile cuts on both plates to cut them out. And then thread the holes (total of 6 1/4-20, 2 #8, and 4 M5)

I’m ordering my screws tonight to get everything delivered on Wednesday and then it’l be take apart the Z axis, which will be fun because one of the button screws holding the MakerSlide is stripped. I’m hoping that I can twist the makerslide to loosen it after removing the other three screws and then slide it out and take off the T-nut.

I’ve got the 269 oz-in motor already waiting to mount on the Z axis. I also need to figure out the Z home switch.

Are the plates excessive? Most likely. Why not just drill the Xcarriage like others? A few reasons but one is the SuckIt mounting on the plate directly is what I wanted and another is that I don’t feel confident enough to drill the Xcarriage accurately. Plus I tend to make things way more complicated because well, that’s what engineers do, right? :slight_smile:

1 Like

I made my own mounting plates as well…it seemed like the right way to go for me as well.

One recommendation I would make is after you have all of the holes cut, use screws to hold down the aluminum so when you do the profile you won’t need tabs and that will save a lot of time on cleanup.

I notice you are on the stock wasteboard so screws may not be a choice you are willing to make. In the future I would use a sacrificial board on top of the wasteboard to screw aluminum into to avoid tabs.

I don’t have tabs in my toolpaths (I live on the edge!) nor do I care too much about my wasteboard…it’s a waste board.

Once I get my Z axis upgraded, I’ll be skimming the board (already have an Amana bit for that I’ve yet to even unbox) and I’ve already cut into it some and had screwed in a corner bracket as a pseudo bump stop. I’ll probably throw some M5 and 1/4-20 screws in each plate just to be safe.

I was just going to rely on the climb cutting and carpet tape but a few screws will save a couple hours of cutting, aluminum, and reduce the potential for injury :slight_smile:

1 Like

I am looking forward to seeing how this comes out. I have the exact same parts and will need to undertake this soon too.

I do like that you have included mounts for the Suck-it rails

Here’s the screws that I ordered from McMaster (which being in Atlanta means Ground shipping cost but overnight delivery…so awesome):

Spindle plate to Z axis - 1/4"-20 5/8" long socket cap with thread locker (here)
Spindle mount to spindle plate - M5 x 0.8 - 16mm long socket cap with thread locker (here)
Linear Z axis to carriage plate - 1/4"-20 Hex drive flat head 5/8" with thread locket (here)

  • I had some 16mm M5 screws and test fit with the CNC router parts mount. I had 8mm of thread sticking out the backside which gives me at least 1.5D of thread engagement which I like. The next size up from 16mm was 20mm (with thread locker) and that would give 12mm sticking through which would exceed the plate unless I put washers.
  • The Z axis back plate is 1/4" thick and a 7/16" plate means you have 11/16" of length (and those screws are measured from flat to end) so going with 5/8" should give me clearance and 1.5D thread engagement.
  • The spindle plate holes are counter-bored approximately 0.3 inches deep leaving just over 1/8" of material. The Z-axis block appeared to be a good 1" thick so will have 2D thread engagement. Ill probably washer these screws so that will also add some thickness.
  • I plan to reuse the Z-axis motor mounting screws, washers, and lock nuts to mount to the carriage
  • I’ll locally source the #8 screws for the dust boot mounting track.

The aluminum plate is ideally 7/16" thick. The reason for the thickness is that originally I was thinking of using some helicoils instead of just tapping the holes and I want at least 1.5D thread engagement. The plate was 6" x 12".

1 Like

Nice!

1 Like

So I cleaned up the thru holes tonight and got them all cut through and also cut the spindle mount holes. Sorry I didn’t snap a picture.

I had to stop for two reasons:

  1. I didn’t have the proper cutout Gcode generated. I changed my plan from cutting all the way through but 0.0005" extra length for a finishing pass to just doing it all in one toolpath and to final dimensions. And I forgot to change the Vcarve settings and save the toolpath.

  2. I needed a Lowes run for some nylon spacers/washers for the stepper mount. While cutting the spindle mount holes, I decided to start tackling mounting the stepper. It appears a 1" and a .25" spacer was included but it still left a gap, as seen below. I didn’t want to

There’s plenty of motor shaft engagement so I figured I’d get a 1/2" spacer to add to the 1" spacer and all would be good. I get back from Lowes, scoot the motor out (still plenty of motor shaft to screw the coupler on), thread the spacers and try to screw the screw in…no threads are engaging. Argh! The screw is too short! Now I could do a few different things now. I can order the proper length spacer from McMaster, buy a longer screw, or buy shorter spacers/washers. I like the look of a single spacer, hate the cost ($11 + shipping). I can get 1/8" spacers from Amazon Prime but then my thread engagement is only going to be around 1/8" which is probably sufficient but not for me. Easiest is to buy 2" screws and be done with it so I’ll probably just go that route. I’ll blue loctite everything once it’s finalized.

I’m hoping that the home switch can be mounted directly on the top of the axis with some self tapping/drilling screws.

This

plus washers to distribute your hold down force over to a larger area.

Make sure

  • your safe Z clears the screw heads
  • your profile clears the washers

I’ll use the holes with a counter-bore for hold down. I have enough on each individual plate that should be sufficient. And since they are counter-sunk, the heads shouldn’t stick above the surface. I might need a few shorter screws and washers for that.

There were some bad storms in the area last night so I held off of cutting. Hopefully I’ll finish the cutting this weekend and get everything tore down and started. I’ve had some semi-minor cutting issues related to the Z axis that I’m hoping this will correct going forward.

I did buy the stepper hardware and I ended up going with a different spacer plan. I found some 3/16" spacers and some 1/8" spacers at Ace Hardware so I’ll use those spacers with the 1" spacer and it should be sufficent. I also bought some stainless socket head 10-32 screws to use instead of the phillips screws that came with the Z axis.

Yea I had contacted Phil and asked but I decided to go a different route, primarily because I didn’t have washers so it was a trip to Lowes anyways and again, I make things more complicated than they necessarily have to be. Just how I am :slight_smile:

1 Like

Did some more today and luckily stopped the machine before disaster. The first few passes went okay but I noticed an increased noise along with some chattering and then saw it start to veer off, hit the pause button and it saved it. Here’s the results:

And the bit:

I think some aluminum welded on but it might have been during the deeper cut. I measured and it was about 0.09" deep but I don’t think I got that many passes so maybe it was cutting too deep (again why I want to change out my Z axis).

I went ahead and did a whole bunch more stuff so I could feel productive and have some wins and thankfully that’s what happened.

First, the stepper mounted with no issues except I’m a dumbass and can’t do math right. My 1 1/2" socket heads were nowhere close to being long enough (which I should have known) so I went back with the Phillips head screws. The spacers worked beautifully and the motor is nice and tight and secure. I decided to connect it up and test it (this is a 3" jog up then down).

I did a 4" jog at 10 ipm and it was just smooth motion. I had to change the steps/mm, $3 and disable soft limits. I’ll need to adjust max Z travel for soft limits later and actually calibrate it and such.

Then I did a dry check of some of the mounting holes (the intent of the pictures is to show the light through the holes):

All of the spindle plate holes lined up and so did the spindle mount holes! Woohoo!

I then took it a step further and got out the metric tap, some WD40, and went to town on the spindle mount holes. Got all four tapped (did a quick cleaning with the proper drill bit before) and also cleaned up the 1/4-20 holes a tad with a 1/4" drill bit (not the threaded, the thru holes). Went and fastened on the spindle mount and it fit great, except slightly angled. It appears the mount holes have about a 1mm slope to them. I should have cut them on the same power cycle as the others but I didn’t and I guess the X axis got a little crooked and caused it, the other holes looked perfect. I checked and I have almost a degree of adjustment when it comes to leveling so I should be good. Sorry I didn’t grab any pictures of this part. I also don’t own a 1/4-20 or #8 tap but a coworker offered to tap them at the office (we have a full machine shop for production and for prototype).

So now it’s decision time on how to finish this. I’ve got a few options, not listed by my current preference:

  1. Resecure the material and cut again but way less aggressive and maybe also with some lubrication. I guess the holes got me cocky thinking I could go aggressive or maybe the Z was cutting much deeper than I thought. I calibrated recently and checked everything before so I don’t know how it gets crazy depths. Again this is a big reason why I’m looking to change out my Z axis, the standard design is just a weak link of the system.

  2. take it to work on Monday and ask one of my machinists to just straight cut out the final dimensions. I had a 1/8" rounded corner in the design but I’d take it being done instead of a round corner.

  3. jigsaw it out myself. Probably a slow and tedious option involving a good amount of WD40. I’d get blades meant for metal and also pickup a tap set so I could finish it this weekend.

I don’t have enough confidence in my CNC skills right now so I’m thinking of the punt option. Plus I should probably get a new bit if I wanted to cut it myself. In fact, this whole thing would have probably been much quicker and simpler manually drilling and cutting but I don’t have those tools and wanted to broaden my CNC skills. It’s definitely been a learning experience.

Oh another note about that spindle mount. I had designed the plate as if it were 2" thick because I looked at a different mounts drawing (they don’t post the 611 mount drawing) but the mount is only 1.5" thick. So there’s a 0.25" lip under the mount instead of it sitting flush on the bottom of the plate. After seeing it and thinking about it, I’m okay with that as I believe my potential upgrade router mounts are 2" thick so those would be flush.

One more thing, I haven’t verified but I’m wondering if my X axis V-wheels might be loose or came loose, hence the chatter.

I just placed an ordered for the eccentric spacers, socket head screws, and locknuts from Inventables. These will be added when I swap out the Z axis. Planning to do both the Y axis and X axis wheels. The parts have been in my cart for months, so it’s time to do it.

I’m going to try to make an Easel project and share it for the plates just in case anyone else may be interested. I’m also willing to share my Vcarve file (and I’m debating about making an STL file using the heightmap program…

You could always just use a hacksaw to cut your plate out. Fire a new blade in and it will go through it easily, then you can file/sand to tidy it up

I’m not overly proud but the cutting job is done, took the jigsaw to it this morning. A lot of WD40, 2 blades, and about 2 hours later:

When life hands you lemons, make lemonade, right? Since the rough edge is going to face up, I had no real need to trim it so I didn’t. I’ll smooth it out with a file though.

So now I gotta run out and get a file to smooth it out and also a fractional tap set to take care of those holes.

Once my Inventables order arrives, I’ll do all of the upgrades at once so I’m not rushing to tap the holes. I trimmed the T-track to size for the SuckIt mounting. While I have it all apart, I’ll probably also check my Dewalt brushes.

Maybe one day I’ll buy another aluminum plate and try this all again on the CNC. For now, just gotta get it up and going.

I decided to take apart the stock Z axis last night. I needed to make sure I could get it off without severely damaging the X carriage, worried that I’d need to quickly place an order with Inventables for a shipment before the weekend.

Luckily it went smoothly. Disconnected the stepper and the home switch and then I loosened the Dewalt mount and just removed the 2 screws from the motor plate and the whole thing came off. Figured that was easier than taking off the stepper. Then after removing the three non-stripped screws, I was able to twist the MakerSlide a bit and wiggle it off and now I have no Z axis:

I checked the Vwheels on the X axis, at least one was loose. Argh! But my order with the new hardware shipped (supposedly) so I’m hoping to get that in tomorrow.

Here’s a photo of the stepper and the spacers (Catbide!):

I’ve also planned out where to mount the Z limit switch (the flat surface below the motor mount plate). I’ll drill and tap 2 M2 holes that shouldn’t affect anything. There’s room under that overhang to mount and not have the wire touch metal. A washer or spacer would also guarantee it and push it out a little so I’ll probably try that. I’m also thinking of tapping a hole into the top of the spindle mount plate to put either a screw or threaded rod to act as an actuator of the switch. I’m thinking a 10-32 threaded but M5 and 1/4-20 are options as well (I have screws already for those threads).

I know I’m long winded but sometimes, laying out the thought process, allows for others to tell you where you can improve.

Me again.

I rasped, filed and sanded the edges and tapped the holes. Not a trivial task. I was surprised by how well the aluminum responded to a sanding sponge. Anyways, after that it was time to test fit everything.

I tested everything except for the linear Z to the carriage and everything actually worked. I should have opened the spindle plate 1/4-20 holes more but they all worked.

The third most important part, do the SuckIt rails work?

Woo! Need to make sure the other dimension (outward distance) works but I think it’ll be fine.

Now the two negatives:

  1. The 20mm screw for the X carriage is WAY too long. 12mm should be plenty but I’ll go with a 16mm because I have spare ones…
  2. The right side eccentric nut/spacer has room to move, the left side, however, does not. I’m going to see if a coworker will take a bit off at work tomorrow.

So close yet so far!

1 Like

It’s alive!

Action video: https://youtu.be/tHy3rpnKSgw

A few notes:

  1. very little mounting slop for the Z axis to the carriage. Kinda sucked but I made it work (for now…gotta skim and see how it looks)
  2. I’ll be able to put Charley’s raised Y plates back on and not have any problems reaching the wasteboard.
  3. The Dewalt sticks about 1/4" out further than the hole in the SuckIt which doesn’t have adjustability outward like that. I’ll put some spacers behind the aluminum channel (ones I had left over from the stepper) and it should solve that issue.
  4. Switching out the eccentric nuts with the spacers was both a pain but also so needed. It’s amazing the difference between the two methods.
  5. homing works perfectly with the switch mounted and it looks clean.

So now I just need to recalibrate all steps/mm and then skim the wasteboard. I upgraded to GRBL 1.1f as well today.

1 Like

This should be my last post on this topic (unless there are questions from others) as I am considering this upgrade and everything that went with it to be done.

I went down tonight and did a few things to wrap it up:

  1. Changed out my end plates back to Charley’s lifted end plates. This new Z axis has PLENTY of vertical motion. I can probably go an additional 1" of lift (2" total) and be fine to carve down to the stock wasteboard.

  2. Rechecked all the Vwheels and belt tensions and had to adjust a few of the Vwheels.

  3. Added the spacers for the SuckIt Mount. Since I was skimming tonight and my 1" bit fit in the hole almost exactly, I used that to align front to back. The 1/2" spacer with about a 1/8" setback was perfect. The other spacer I tried (can’t remember the size) was almost enough but not quite enough so I had to go with the 1/2". Had to wrangle the left/right position of the track a little as well but primarily because of the 1" surfacing bit.

  4. Skimmed the wasteboard. I was severely worried that it was going to be horrible however, it was just fine. Had some slight ridges in the Y axis direction only indicating a slight out of tram condition. Nothing major so I’m not going to worry too much over this. I’ve still got some of my original aluminum so I could create a new spindle mount plate that could allow for more tram adjustment (or better relative alignment to the mounting holes). I actually hadn’t calibrated my steps/mm again after the Vwheels and belt adjustments but I was doing a 750mm x 750mm square for skimming that was zeroed on the stock wasteboard grid. The skimming went right on the grid lines.

So now the wasteboard is skimmed and I’m ready to go back to carving. I should verify the Z steps/mm but that’s simple and easy. Good thing I’ve got a 4 day weekend coming up!

1 Like