Z-axis drilling/plunge is the only place I've seen browning/heavy friction

lets make sure we’re using 100% of the existing motor, and have made all best attempts to wrest out the slop from the router base before moving into additional z-motors.

there are plenty of 3d-printer ppl here (though I am not one ) to get a higher tolerance model of the screw-router block put together. that and the connection at the bottom of the screw (outside the housing) are the sources of most of the slop I’m seeing.

1 Like

Aren’t these two different issues? As in, the feed rate of the base is related to the motor speed/lead screw pitch, and the slop is related to the build quality of the depth adjustment assembly?

While I do have a 3D printer myself, it’s not a terribly good one and building a Prusa I3 to replace it is on my to do list. I wouldn’t (yet) describe myself as one of the 3D printer crew.

I saw a thread a little while back where someone managed to shim the slop out of the lead screw. If I’m remembering correctly, they used a retaining clip and some washers until it didn’t have any appreciable backlash in it. I’m going to have to do some digging to find the thread though.

I’m planning on transitioning to a more traditional Z-axis for my machine for more reasons than just slop and feed rate, but they’re the big ones and the most frequent reasons I cite.

1 Like

I’ll look into these points this evening. Thanks!

yes, there are 2 issues z-Axis speed and drive gear sloppiness. I should be able to compensate for it in Fusion but not quite in today’s cuts.

The main reason I wanted to keep focus on the Default Setup is to make sure we’re exhausting what can be done with it before moving on to bigger possibilities. I encourage everyone focused on MaslowCNC improvements to try the same. It’s only as I type this that I realize I’ve completely abandoned the wood frame, so… heh.

Anyway, I’ll take a look through the old posts, I think i remember that too from months back.

1 Like

The problem with the Z axis moving slowly isn’t anything to do with tolerances,
it’s just a matter that the very high gear ratio of the motor is going to move
slowly, even at max power.

the only way to fix that is to have different gearing (or change the voltage),
and it’s far easier to switch to a different motor with the same physical size,
but different gears.

If we then find that the faster motor isn’t powerful enough, we can switch to a
bigger motor, but we should test different gearsets first

There’s nothing that can be done with the default setup to speed up the Z axis,
but if we can test different motors, the next batch may be able to go out with
different motors (but we need to test early, as it takes a long time to order
motors, and we don’t want Bar to have a pallet of the wrong motors on hand.

That’s pretty sound reasoning. From my initial sketches I went with a welded steel frame, so I’ve never really thought of myself as being anywhere near the “stock” setup. As a result, I feel the need to constantly improve and innovate, even if it deviates from the original design. Maybe this means that some of the lessons that I learn from experimenting could be applied to the stock design?

If I could do this again, that would be a pretty good mindset to follow.

Would it make sense, then, to play with a couple of pulleys first and try to get the motor to stall to get a good approximation of just how much torque is needed from the motor? A couple different size pulleys sounds cheaper to me than a set of different rpm motors. Maybe it wouldn’t give us as controlled results, though, for the stock setup…

I wish I could get this testing done myself so I’m not simply a backseat driver… Really need to get some time off from all this moving to play with the Maslow again. Unfortunately, that might not be for awhile. :frowning:

1 Like

I’ve started thinking about this too.

If I’ve read the forum correctly, the present 12V motor gives 11kg.cm of torque on ~.5A doing in the neighborhood of 12rpm at the shaft.

I’m thinking of this one, which lists 72kg.cm of torque and ~50rpm. It is thirstier, though - a smoking 6.5A stall current instead of the 2.4A of the stock motor. That would fry a stock Maslow board in an instant… It needs the connector signals re-arranged.

1 Like

I think we determined that these are the motors

available in 12v or 24v and in many different gear ratios resulting in

6,10,15,18,30,38,55,88,`50,165,265,500 rpm

although double check that all gear ratios have the same dimensions

$24 each

the one you link to lists a stall current of 6A, which will fry the system
(unfortunantly the one I link to doesn’t say it’s stall current, but it’s no
load current is only 45ma compared to 250ma for the CQRobot one)

1 Like

I disagree that ‘nothing can be done’ about speeding up the Z-axis motor, at least until we know we’re burning motors or shields. So far, we’re (I’m) just a little frustrated that its working so well, and could be working faster. =)

Today’s progress aside from Z-axis slop and speed constraints:

I was able to get some excellent ‘constant loading’ on long cuts with continual z-axis movement, apologies for misidentifying it as an issue, it was entirely my Fusion360 settings.

I was able to get my head around Fusion’s ramping features, mostly loosening up everything away from presumably metallic milling settings.

Similarly with tabs, and other Z-critical things given the slop in that axis, i ended up making much larger tabs and over-cutting some more into the wasteboard. Excellent entry and exit cuts with very little tear-out. (again on 3/4" home depot cabinet ply (the best looking pre-sanded ply in homedepot, sic)

2 Likes

The torque rating on that motor is 9kg.cm at 24V, so probably half that at 12V. That’s less than half the stock z-axis motor. Higher speed and lower torque doesn’t seem desirable. Still, if someone tries it, we will know.

we are applying full voltage to the motor, and it’s moving at the rated speed.
what else can we do to speed it up?

Gearing, or a belt drive:

we try to fit a small gear on the router base screw and consecutively larger gears until we can’t turn the screw any more with the motor, start losing motors, or blowing shields.

which leads naturally to the possibility that more power will be needed (could be utilized) in the future.

I doubt much work been done to see where the next appropriate level of ‘upsizing’ would be on a future Shield board, but perhaps that might be worth beginning to investigate…

or a ‘power agnostic’ shield, that doesn’t require a motor’s direct current into the shield?

or some sort of stacked ‘shield per axis’ setup?

I’m thinking about that kinda current (per @blurfl 's link) per axis and you’re talking real amperage… not that X-Y needs more power, but thats a whole other thread…

My point is to begin the discussion of how best to prepare for the possibility. .

This looks like the best info on reducing slop in the router base: Z axis only moving the router some of the time

That’s certainly the post I was thinking of. Thanks for finding it, I spent most of last night working through designing built-in’s for the new home with the wife. I clearly mis-remembered how the fix was done, though >.> Not sure where I got the idea of washers/retaining clip.

Also, super cool to see @blurfl approaching more powerful motor drivers in this thread! It’s really cool to see the progress happening!

Based on your link in a previous ‘speed up’ thread I’ve got a set of larger gears for the X-Y motors, I’m holding off on them until we get some machines really humming well and I’m more comfortable with really tuning the Fusion360 paths.

To that end I started a Spreadsheet here for Feeds and Speeds observations/experiences and a detailed checklist for Fusion360 CAM settings, so I can begin to build up a library of Presets.

I’m currently using McMaster bits as they seem easily replaceable and, at least in the USA, immediately accessible to everyone.

I’ve had some excellent cuts with no tear-out on the entry at all, and I’m going to try to layout CAM operations to do ~60-70% cuts using that and flipping to an up-spiral for the finishing operations.

I’ll get a set of Maslow branded bits once they return to full stock.

3 Likes

Do you guys think this would replace the stock z axis? My options are a bit more limited being in Canada.

I need a bit more speed and this one looks to be high enough torque but not crazy amperage draw.

Thanks

possibly, note that the stall torque of this motor is 3A, that’s enough to blow
the stock motor controller chips

2 Likes

Would you mind starting a new thread for this? The detailed results need a clear home separate from the more…theoretical?..planning we’ve been doing in the Max Feedrate thread :slight_smile:

1 Like

I’m happy to but it’ll be a few weeks (2) before I’ll be able to post my results, I’ve got some travel coming up this week and next.

I haven’t yet had the chance to test everything and make sure i’m not wildly off. (I don’t think I am, based on my successful cuts so far, but I want to test again before sharing)

In addition to my upcoming travel, I have some upgrades to make to my Sled to incorporate some of the solutions to the Ridgid screw-slop.

I’ve plowed through most of the relevant 2D-Operations in Fusion360 and made up a separate sheet for each operation including all the tabs in Fusion’s CAM processes. It was quite enlightening, and helped me get my head around each operation, and more importantly, some of their setting’s subtle differences.

I recommend the exercise whole heartedly!

2 Likes