This is a good time to bring in accuracy vs precision
the old maslow was very precise, you would get the exact same cut each time. But
it’s accuracy was not as good, especially towards the corners
maslow4 currently is working on fixing/improving the calibration to be more
repeatable. But is expected to be more precise and more accurate. It’s already
more accurate, but the calibration issues are limiting it’s precision.
David Lang
How detailed are we talking…2mm letters or that the smallest detail in the letter is 2mm tall? 2mm letters is a big ask, but having a 2mm detail is no problem at all. I’d say that a 1/8th inch bit is the smallest I’d really recommend but that’s mostly a speed issue. With a 1/8th inch bit you can only really cut about 1/16th of an inch per pass so we’re talking like 12 passes to get through 3/4 plywood which is pretty slow.
i have always appreciated that graphic. It gets right to the point.
Correct 2mm detail. It would be cool to get 2mm letters size though. I would be doing this on luan <1/4" so 4 passes isn’t terribly daunting…
Another thought is maybe add a clamp frame that can fit onto your cutting frame. pressed on either side of the fabric and clamped together could keep a fabric from being stretched but held in place. If you line that up with the cutting surface of your frame so the fabric can rest against that base surface, that should provide plenty stability, rinse and repeat for other sections if it’s even larger of an area done at once maybe? Or maybe your base surface when cutting is high friction cutting surface or Sintra.
I guess I’m doing it wrong then, but I cut 1/2" ply all day long in 2 passes with a 1/8" bit. I found that only doing 1/2 the bit width for depth dulls the very tip and never uses any of the rest of the cutter. If the bit is sharp, it cuts well and will progressively get louder as it dulls. Once it is dull, it will fray in the middle pywood layer and not cut downward or out in the lower corners and then it is time to put in a new mill bit. I use up a bit every other sheet roughly as a consumable cost but I buy them in 10 packs for $15. I expect the maslow 4 will use the bit faster, but the corner issue will go away and the noise will be the deciding factor.
My experience is limited to about 10 sheets cut this year, so I’m not full time production or anything and my cuts don’t require extreme precision either, but I think the conservative recommendations are quite conservative… for good reasons.
depending on what you read, there are recommendations to go 1/2 bit width or 2x
bit width. Bar tends to be very conservative in cutting depth.
A lot of it depends on how ridgid the machine is and how much force you have
available to move the bit.
David Lang
Are they just HSS? Sounds like you found the sweet spot in terms of speed vs. cost.
They aren’t the top of the line for sure, but I’ve used these ones. The claim is that they are tungsten carbide…
“toung’s tounge car-byde” amma right?
Will this be available for sale after the Kickstarter and at what price?
I am very Interested, but I don’t suspect I will have time and space to use this until next year perhaps. Also, as much as I would like to be part of the community and improve this system as an early adaptor, I have young kids and don’t have much in the way of spare time at this point in my life.
in the past, they have been available after the kickstarter at the same price. I
expect the same this time.
David Lang
We will make them available after the Kickstarter, but when the Kickstarter ends our main focus is going to be on delivering to our Kickstarter backers so we may not open orders again until we’ve delivered to all of our backers or at least until we are close to delivering to everyone.
Would creating a crude (or dummy) rail system overcome some of the large cut surface balance issues with the sled model? I’m asking this out of genuine curiosity. What I am describing is a two or three rod&rail track that is bearings only, no motors, connecting to the sled on the x and y axes, while keeping the four belts mounted as intended.
This would allow the sled to float on the rail rather than depend on level surface. If the current M4 setup seems to have a good overcoming of resistance as shown in the videos (cutting the screw, hitting a knot), then having a reasonably low friction rail could keep the sled mostly plumb. The third rail I’m thinking is a second z-axis to correct the majority of tilt anticipated. And I suspect because the rail doesn’t have to be fixed, or have any cable/motor/precision adds to make it costlier
This is more in reference to alternative uses like painting/sprayer/marker/glue. Even if accuracy is reduced for the zero-project-contact situations I’d argue it would be well within acceptable tolerances for the medium. And certainly less expensive than fully building a dedicated rail-motor system.
Speaking of which, have people found any success in buying standard bearing sizes from Temu or other cheap-ship stores? Sometimes there are gems amongst the junk…
once you go to the trouble of building a gantry system, you are better off
building a coreXY system, it’s more accurate, better understood, uses fewer
motors, and has very wide support in controller boards.
(this probably needs to be added as a FAQ)
David Lang
I’ll pile on with the follow-up FAQ, which is, “with all those advantages, why go with the Maslow at all?” to which I would reply that the Maslow is desingned to be simple and low cost. A full gantry system is neither.
especially a gantry system for a 4x8 ft work area. To cove that span without
measurable sag requires very beefy frame elements (that then need more beefy
components to support them, and large motors to move them, especially to move
them fast with high acceleration). that means that just the metal for the frame
will cost you more than the maslow, before you consider the motors, movement
components like wheels, electronics, machining of the frames, welding, shipping,
etc
David Lang
To be clear on my point from above, I was hesitant to call my rail idea a ‘gantry’, since it’s more stabilizer rods on a bearing and rail that can move smoothly in 2 axis passive motion. No additional motors would be needed, since the M4 would still be anchored as it would in the corners. The sled would be supported by the rails, yet directed by the belts. Imagine a high rise obstacle course where the user is harnessed on an overhead track in case they fall. In this case it’s not a rope hanging, but bars for a stiffer z-travel lock. It would be a way to get close accuracy for gluing without dragging the sled, and also not breaking the bank with materials and tech. If I have enough laying around the farm for a proof-of-concept, I’ll try to make it in the next few days.
I believe something like you have described was proposed a few years ago for the Maslow3. Im not sure how it turned out but there were pictures if I recall. might be worth a look for some ideas
the issue with the gantry isn’t moving it (you can easily move the gantry with 2
motors vs the 4 motors the maslow uses, and with more accuracy and precision)
it’s with the gantry supporting the weight of the sled/router without deflecting
from it’s weight.
Once you can build a gantry that can hold the weight, then you can add motors
and belts/chains and make a corexy drive system.
https://corexy.com/theory.html
David Lang