So I’ve been in and out of the Maslow forums for the last 6 months or so.
I’ve gotten some awesome response feedback from @bar and other seasoned Maslow users like @wouldchuck@dlang and many others.
Here’s where I’m at with my Maslow:
I got the anti static hose because the static kept shorting out the Maslow board and resetting mid operation.
Switched from using MakerCAM to Carbide Create, drastically improving the outcome of my g-code.
Built frame extensions around my spoil board with attachable tops to match the material I’m cutting, preventing the Maslow from tipping when near the edge.
Our Maslow is used for the Sign Shop I work at. We often go straight to using an entire 4x8 sheet of material.
The biggest issue I’ve been running into is the Maslow losing accuracy after its first pass, resulting in a botched project and the loss of entire 4x8 material.
My theory is that because I’m often going to the edges and corners of the 4x8, the Maslow loses accuracy due to my frame being the standard size suggested when we first purchased the unit.
With that being said, on top of needing suggestions for the frame size to get consistent accuracy, my question is: What did you change or modify with your Maslow or set up to get your unit running efficiently and consistently?
I’ve had more failures than successes since obtaining and operating the Maslow that I’m more terrified to begin a project with it (Because of the massive material loss with each 4x8 project, and time) and I really want to get the machine humming.
Open to all suggestions. We have a smaller 3D printer available and can print parts if needed. I had @dlang suggest some better towers and guide-rod hold downs.
At this point, I want to sit down for a few days, take the machine apart, make all of the upgrades and modifications, and get the machine up and running. Upgrade the frame too!
Thank you in advance! This forum has been great to refer to when I’m having troubles, especially as someone who is not from the carpentry/mechanical world (I’m a salesman/graphic designer who had his boss throw this deconstructed machine at me and say “Make it happen”)
There is always a lot of good work happening in the firmware development here. That said If I were trying to run a machine for work in a stable environment I would pick a release that worked for me and stick with that so at least I would know what works. It is easy enough to go back and forth to jump forward to new firmware and then go back. I like the new find anchors system in 1.24 and I like the trace outlines system but I had a lot of success actually cutting with version 1.17 I guess the advice is that once you get it set up again and working that I would pause and document for yourself a good yaml file and a good firmware version.
I haven’t had any problems with internal accuracy once it gets started. I would make sure that your anchors are free to rotate on their anchor pins.
I have been having fun with frames but really the most solid and reliable is probably drilling a floor if you are allowed to do that. Large frame setup if you can. Mine has been happy with the 4 by 8 sheet with an 8 by 14 foot frame.
Take time to do dry runs and practice runs with your bits on trash wood so that you can get reliable results before switching to the good stuff. Hopefully once you get it set up, you can skip these but getting started it is worth doing a lot of practice things first.
Check your clamps and cylindrical supports in the maslow. All of mine are breaking at the screws. The turtle clamps and printable pillars look good. People seem happy with them.
As you get it set up write down what works and keep a log so that you can systematically go back and change one variable at a time when tweaking your system.
Hope that helps, keep us informed and best of luck.
I’ve only really used the 1.24 firmware since assembling the machine. I also like the Find Anchors function, I’m unsure about how to use the trace outline system however.
I took note of this and made sure my anchors have lots of room on their anchor point!
I honestly have yet to take the Maslow apart to make any replacements or adjustments. It’s hard to dedicate the time to doing that when work is already so busy. But I’ll take note to inspect those pieces when I finally do, and I’ll probably abandon lock nuts as someone has suggested in a different thread, making disassembling easier when the time comes.
Yup, been taking lots of notes on all of my different operations. When the machine fails more than it succeeds, not taking notes would drive me crazy trying to diagnose the issues.
I chose these dimensions based on the frame calculator and found that the workspace was basically green lit, so I’m hoping I get more accuracy going forward. I mapped out the extension pieces in my design software, cut the proper angles out of 2" x 4"s, screwed them together, then fit and screwed them to the corners of my existing, initial frame build included in the start up guide.
Since I often cut 4’ x 8’ material, I added some removable, adjustable bumpers to my workspace. The idea is that I’ll change out the bumper material based on my project material’s thickness. This helps my material from moving around and also prevents the Maslow from tipping when it reaches the edges.
The two problems I recently had to deal with, basically simultaneously:
My Dewalt Collet kept loosening and dropping my bit. This is a bit I’ve used before, and I even loaded a new bit. Not a fun sound when the bit drops out! The sparks also weren’t very encouraging. I checked my bit size and it wasn’t an odd size, so it must be the Collet. New one on the way from PreciseBits.
I had multiple objects I was cutting from a single sheet. The first object would finish cutting (or so I thought, but it didn’t cut all the way through) and move on to the next, but when it went to start cutting, the Z-Axis didn’t drop. The Maslow kept moving, and ESP32 said the Maslow was lowered into the material, but it wasn’t. This happened twice after a restart. The fix was sending more amps to the Z-Motors. 1.4A Run Current and 0.75A Hold. This instantly solved my issue.
Just another day of me battling with the Maslow Unfortunately the Collet broke just as the machine seemed like it was finally working, and now we’ll have to outsource the routing since we have to wait for the Collet to come by mail.
trace outline is a button that you can push right before cutting that will move the machine around the edge of the design area. it is abutton on the main screen
Can you expand on this comment? Like are you doing cuts in different z heights and they aren’t lining up? Maybe the material is sliding around ?
I just looked at your frame and the extensions. Are you able to screw those down to your floor, they might be flexing or moving on you that would accidentally change expected positions.
I was having issues with it before I added the extensions. My best example is I would be in the middle of a run that was supposed to use nearly an entire 4’ x 8’ sheet for a single object (A photo frame for a kids camp was the last one) and It went around the entire object without issue the first pass, but as it would work on the second pass, it wouldn’t be accurate to the existing pass it made initially. It would veer off, making two lines, and the Maslow would eventually E-Stop because it could detect it was out of place.
I had the material fastened so it couldn’t move.
I’ve thought about fastening the extensions to the floor as well, but I want to see if I do end up having flex issues first. The frame might not stay where it is so I don’t want to screw into the floor at the moment.
Experimenting with a Laser WebUI on Our Modified Maslow 4.1
I wanted to share something we’ve been experimenting with on Laser Stryder, our heavily modified Maslow 4.1-based development machine. MR2QUIK is the developer behind Laser Stryder, and the machine has evolved quite a bit from a stock Maslow—with a different sled, belt and motion-control hardware, encoder setup, and plenty of other changes. Because of that, this isn’t intended as something you can simply load onto a stock Maslow 4.
What I thought might be interesting to the community is the WebUI work. We’ve added a small Laser Spot section directly into the Maslow interface with adjustable power and pulse duration, along with a FIRE SPOT button that only becomes available once Maslow reaches Ready to Cut. We’re currently using GPIO48 for PWM and moving the machine to a 20T encoder pulley configuration.
Earlier Maslow-Mod. motion testing that eventually led to the Stryder platform. This was already being used to evaluate large-format XY tracking before the current laser-control development.
We’re also working with a modified [FreazyKam] (IArchi · GitHub) workflow as the CAM side of the project, with the goal of passing laser power, speed, and operation parameters into the Stryder WebUI while retaining Maslow’s existing machine-state and safety controls. Our development fork: Mii3D/FreazyKam GitHub - Mii3D/FreazyKam: SVG to CNC gcode converter · GitHub
There’s still plenty of testing ahead, but I wanted to share where we’re headed and get some input from the Maslow community as we continue developing Laser Stryder.
Not primarily. The laser is just one of the applications we’re developing and happened to be the focus of my latest testing. Stryder is really evolving into a large-format, multi-tool CNC platform. We’re continuing development around routing, laser operation, the sled and belt/encoder geometry, motion accuracy, WebUI, and pre-cut verification. A lot of this grew out of experimenting with the Maslow 4 architecture, so I’m sharing it here because the experience and feedback from longtime Maslow users has been extremely useful. The laser work is just one piece of the larger Stryder project.
I had cut some large things in 1.17 or before with no problems with self precision on many passes. I also had problems with it recently drifting and stopping itself at 1.24. two thoughts. I might let it run a large dummy program of a rectangle or large perimeter cut with no bit and multiple passes to check behaviour before my next cut. I also might measure anchors and diagonals by hand to check the accuracy of find anchors. I wonder if sometimes the new fast find anchors is returning an off result. When I ran it on a range of frames some of them came out too short in the y.
I’m always around to bounce ideas off, but overall I just want to say that this looks fantastic and is exactly the type of thing that I was hoping to encourage with making it open source. This looks fantastic.
Unfortunately the Collet broke just as the machine seemed like it was finally working, and now we’ll have to outsource the routing since we have to wait for the Collet to come by mail.
