Thanks that’s clear as day!
1 Like
Thought an in-progress cut pic might come in handy at some point. Good luck!
Btw, I was able to calibrate at 2300x1100 with some additional 2x4 blocks to position the corners further out a bit.
1 Like
Very handy indeed. I’ll evaluate more when I’m back in the studio with maslow tomorrow. I’m in a not huge space with it, so I had to shrink the frame a bit and don’t have any room to extend, but I do need more dimensional accucracy than I’m currently achieving (interlocking parts).
I think I may need to see if I can take it out to the 'burbs and use a friend’s driveway…
1 Like
In preparation for experimenting with chopping the clamp, I’ve been wondering about how the arms are intended to fit together? When I build the clamped arm assembly, the arms seem to naturally have some space between them. Is this intentional, or should I be pressing the clamps toward each other while I tighten them so the arms have less play in the Z? My arms are loose enough that they do tilt slightly when tensioned…
1 Like
There is intended to be some play there to keep things from binding up
1 Like
Bar wrote:
There is intended to be some play there to keep things from binding up
I purchased some sheets of teflon with the intent to cut out rings to fit
between the arms to tighen them up without them sticking, but have not yet cut
them out (I will look at doing so when the 4.1 kit arrives)
David Lang
2 Likes
You’ve probably already seen this post Loose belts during calibration - #39 by Andith but I thought it might be worth linking in context of your tilt issues…
I switched to (top down) BL → TR → TL → BR at the New Year and things massively improved with Maslow cutting clean lines nearer its edges. I’m not sure how pragmatic the change in arm order would be for you with the ear removals, that is, if the additional space would be available in a useful space…
I did see this back when it was posted, but it’s much more relevant now. Thanks for the reminder! I was going to try swapping the order of the arms because I’m getting a ton of sled lift on the left side of the frame due to TL being the highest arm and BL the lowest, so I’d imagine this will fix it.
Do you see much lift with the order you’re running?
Also, “top down” you mean starting at the control board down to the sled?
Yes.
I was getting jiggly cuts near both horizontal and vertical edges, and they seem to be gone now. It’s been such a huge improvement.
Before:
After:
Sorry, I know that’s an apples-to-oranges comparison… The after photo was near the bottom edge and would not have been clean previously.
1 Like
I also flipped the Alternate dust port design in my slicer before printing so I’m not whacking it anymore, which had to have been effecting my calibrations.
The wobbly cut looks like a different problem entirely. My guess there is that something happened while cutting. With the ethernet cables I had the signal drop and the machine lost track of where it was causing something similar.
Mine will cut very cleanly close to the edges, but the X axis was skewed by about 10mm near the edges. I’ve got that down to about 2mm now.
What I was having a lot of trouble with, and rearranging the arms will likely help with is the bottom of the sled lifting on the left side when traveling up. That’s because TL is the highest arm and BL is the lowest so when TL is pulling the machine up it has more leverage than BL as seen here. This was causing lots of other issues on the left side of the workpiece.
1 Like
The wobbles seemed to be related to the sled hopping on the wood… Which seemed to be worsened by the leverage issue with the arm order… I’m on a less wide frame because it’s all I can fit in my studio, so the edge issues show up more strongly for me 
I couldn’t make the frame wider without blocking the door to the adjacent studio, and I didn’t want to have to get on a ladder to hang the Maslow if I had rotated it 90º.
When I changed the arm order, I went from 15º lean to 22.5º and waxed my sled. I happen to be sitting next to it at the moment, so I just flipped it over and thought the wax wear pattern was interesting… it seems to show more pressure on the BL corner, where my top arm is located.
1 Like
Will Puckett wrote:
The wobbles seemed to be related to the sled hopping on the wood… Which
seemed to be worsened by the leverage issue with the arm order… I’m on a
less wide frame because it’s all I can fit in my studio, so the edge issues
show up more strongly for me
making the bottom of the sled smoother will help (low friction tape, waxing the
sled, etc)
David Lang
1 Like
Going back to that thread, where I ended up after some testing was this:
- For vertical-ish frames the principles guiding arm layout are:
- Opposite corners should be next to each other in the stack of arms, i.e. the pairs are TR<->BL and BR<->TL
- The ‘highest’ and ‘lowest’ arms in the stack should be separated by a long edge, not a short edge. i.e. TL ↔ TR or BL ↔ BR
And from the above the ‘best compromise’ order I could come up with was:
- For vertical-ish frames I’d go with the layout of (starting topmost ‘down’ to the sled) BL → TR → TL → BR
With changing the arm order I would strongly recommend redoing calibration. BUT BEFORE THAT, change the Maslow_**Z values in the ‘stock’ maslow.yaml to correspond to the changed order. And use it to wipe any existing ‘maslow.yaml’ file on your machine.
For example, here each value is 22mm different from the next (@bar can you confirm the offset is 22mm?), and they are reordered (by their relative height offsets) according to the stacking order. Please do NOT use these specific values, I’m showing them here for illustrative value only, instead figure out one of them for your specific machine (I’d recommend blZ), and then the rest will be simple offsets from that.
Maslow_blZ: 34.000000
Maslow_trZ: 56.000000
Maslow_tlZ: 78.000000
Maslow_brZ: 100.000000
PS. once you redo calibration these values will end up in different places in the maslow.yaml file on the machine.
PPS. you can also find these values in the code:
- For the front end
ESP-WEBUIthey’re insettings.jsas part of theCONFIG_TOOLTIPS, a guide for users, and - For the back end
FluidNCthey’re inMaslowConfig.cppasdcM4ZAxis, which are the defaults for these values before the machine has loaded up themaslow.yamlfile itself.
4 Likes
Aloha,
Is this a simple measurement? From/to what points/pieces?
My machine is coming apart after my current project so I can install my 4.1 upgrade, and I hope to incorporate this when I do.
Mahalo,
David Negaard wrote:
Aloha,
Is this a simple measurement? From/to what points/pieces?
the difference from one to the other should just be the thickness of the arms.
the actual measurement is how much Z distance is there between the belt at the
anchor and the belt at the arm (you can measure top to top, or bottom to bottom,
etiher eway works)
David Lang
1 Like
Exactly what @dlang said. That’s what I did.
Two things:
- Remember to do the calibration without a tool bit, and with the M4 fully bottomed out on the Z-Axis - i.e. at the ‘Z-Stop’ position. No need to actually set ‘Z-Stop’ as the calibration process will do that for you.
- I would recommend doing the calibration with the full stack of boards that you will typically use. For example, if you use a waste board underneath whatever board you’re actually cutting, then repeat that for the calibration. Ultimately you want to get the same thickness as most of the jobs you intend to do. You may need to add some other offcuts around the perimeter so that the M4 won’t tip over if it gets close to the edge (of any top board) while calibrating, which will likely occur if you do a 9x9 grid calibration.
a. This stack of ‘my typical setup for jobs’ should be included in the measurement you come up with for the Z offsets. So, you’ll want this value before you do the calibration.
1 Like
This seems like it makes the most sense to me. Then along X and Y axis the arms along those axis are only 1 apart, versus currently on the left side Y axis travel is two arms apart. (not sure if that phrasing made sense.)
I would recommend doing the calibration with the full stack of boards that you will typically use. For example, if you use a waste board underneath whatever board you’re actually cutting, then repeat that for the calibration. Ultimately you want to get the same thickness as most of the jobs you intend to do.
I have been thinking about this, would it suffice to change the Z values in the software to adjust for the workpiece after calibrating on the spoilboard? If that would work, I was going to recommend a setting on the config tab to enter in the workpiece thickness.
2 Likes
Kyle wrote:
I have been thinking about this, would it suffice to change the Z values in
the software to adjust for the workpiece after calibrating on the spoilboard?
If that would work, I was going to recommend a setting on the config tab to
enter in the workpiece thickness.
Yes, there are plans to add a parameter that you can set for the workpiece
thickness.
with normal plywood it’s not likely to make a big difference, but it can make a
difference.
David Lang
1 Like
I did some measuring with the calipers this weekend. To confirm, the offset is 22mm. It seems the above numbers are from the arm to the bottom of the sled? So these are the numbers that should be used when the arms are mounted at the same z height as workpiece?
I’m really dense in the thinking sometimes (most of the time), so apologies if this has already been navigated. It seems, as David has mentioned, that using the heights from each arm to the bottom of the sled as one set of variables and then appending a fifth variable to capture the distance from the bottom of the sled to the frame mount would be useful, although I know some people have also experimented with varying the mount heights…
I also cut a ring of 3 mil acrylic over the weekend. I had been concerned that perhaps the amount of play above the top arm was problematic in the dimensional accuracy arena. I installed the ring above the top arm, reducing the space from a measured 4.5mm to measured 1.5mm and reran calibration.
My largest issue at this point is that I am cutting parallellograms—if I cut two adjacent squares and place one over the other, they line up beautifully. But when I rotate one of them 90º, the sides are no longer functionally parallel to each other.
This issue persisted after the addition of the acrylic spacer ring. Overall, the ring seems to make the bottom arm slightly more rigid, but overall, the arms all still move well.
1 Like