Inexpensive ways to test and tune the Maslow?

[Rich_Morin]

I’ve been reading about the travails that some Maslow users have had while getting the software and hardware to work as desired. A lot of time and effort is involved, to be sure, but there’s also a substantial cost: buying and trashing 4x8 sheets of plywood and MDF for this sort of thing sounds like a real money sink.

So, I’ve been musing about inexpensive ways to test and tune a Maslow. Because I haven’t received my M4.1 yet, this is mostly SciFi. That said, here are some notions for discussion…

cardboard re-use

All sorts of items are shipped in cardboard boxes. Unsurprisingly, the boxes for large appliances (e.g., refrigerators, water heaters) are very large and pretty solid. Once the appliances have been delivered and installed, these boxes are no longer needed.

So, what about contacting some local firms and offering to deal with some of these boxes? Cut them into sheets (e.g., with a box knife), then mount them on a spoil board and cut them up. When this is finished, cut up and recycle the remaining pieces.

drawing, rather than cutting

With a suitable marking device and holder, a Maslow could act as a flat-bed X-Y plotter. Rolls of paper are fairly inexpensive. For example, Amazon has 100’ rolls of 48" wide paper for $34/roll. And, if there’s a web press operator in your vicinity, you might be able to buy (or scrounge) some “roll ends” from them.

camera-based observation

With a suitable camera, a Maslow could act as a flat-bed image scanner. I realize that this isn’t a quick or easy approach, but since I’m already thinking about attaching cameras to my Maslow, I may give it a try.

My current notion is that I would acquire a large, detailed drawing in both digital and printed form. Then, using the Maslow, I could scan the printed drawing and compare the results with the digital version. By fiddling with feed rates and such, I could learn a lot about the Maslow’s X-Y positioning accuracy, behavior, etc.

any other ideas?

I’d love to hear about other ideas on this topic, feedback on real world experience, etc. Nu?

[md8n]

@Rich_Morin

drawing, rather than cutting

This is definitely something I want to try. Especially as it should be easy for me to swap out my spindle with a suitable insert to hold a marker pen.

[Betalombo]

@Rich_Morin

drawing, rather than cutting

It’s true, I really want this. I would like to draw directly on the wood with some easy-to-clean material so as not to use paper. I was thinking about these 2mm [pencil] refills (Amazon.com)
Inside a 1/4 inch metal casing, driven by a soft spring that does not throw the graphite tip far when Z reaches the safe position, a spring that has a minimum travel of 5 mm

[James_Presnail]

this thread had some 3d printable options to try out

Maslow plotter, or getting used to Maslow without spoiling wood

Thanks for the suggestions! It does fit without the pen holder screwed in (I didn’t bother reprinting after checking the model in Fusion and my measurements match your design). I printed with standard supports in Cura - the tree support generation wasn’t consistent so I used the standard ones for this. I’ll keep working on the threads - you can see in this pic they came out a bit like a ratchet gear on my printer, I’ll take a Dremel to them and see if I can get it working smoothly. Hopefully the…

[Rich_Morin]

That’s a great thread; thanks! FWIW, I also stumbled upon this thread:

Pen or marker use

Using marker or pin on Maslow4? I saw videos just a few minutes ago about using a pen or marker on a cnc to draw path to make a drawing or to check tool path.i am sure the kobalt router could hold marker without much difficulties. Thinking dewalt would have to have an offset for it to work. Being new to this i am thinking this might be useful. Wondering what others think.

[N8DGR8]

Most of my trial and error requires real wood. IE bits, speeds, etc. your best bet is to find cheap wood. A local builder gave us several large pieces of scrap.

Low end plywood is $20 a sheet.

[Rich_Morin]

ChatGPT says that there are quite a few sources for downloadable test patterns for 2D CAD. Has anyone been trying anything like this?

[dlang]

Rich Morin wrote:

ChatGPT says
that there are quite a few sources for downloadable test patterns for 2D CAD.
Has anyone been trying anything like this?

the problem isn’t the pattern, the problem is the cost of the material that the
testing can chew up.

test patterns designed to stress test the X and Y axis of traditional machines
independently, aren’t going to be doing the same thing on the maslow (since the
motion system is so different)

sometimes you can use the same plywood multiple times if you shift it
(left/right/up/down) so that the new cuts will hit different areas of the wood.
This isn’t perfect as the cuts can cross over old ones, so it’s not exactly the
same, but it’s pretty close

you can also do cuts at different depths (first try, set the Z height so that
you just barely make a mark, then in later tests, cut to your normal depth)

sometimes foam boards are going to be cheaper than plywood (and you aren’t going
to dull your bits on them while testing)

But at some point, testing needs to be done with the final material, as it’s
probably going to be harder to cut than the cheaper stuff.

David Lang

[James_Presnail]

Why are you looking for test patterns? Are you just looking for some gcode to test? If so krabzcam is amazing and simple IMO

[Rich_Morin]

A given user project may or may not reveal a problem, but a set of Maslow-oriented test patterns could give folks an organized and reliable way to characterize, report, and reproduce issues.

[Rich_Morin]

Here’s a variation on your idea…

Some years ago, I learned that whiteboard material is pretty cheap:

32 sq. ft. 96 in. x 48 in. Hardboard Thrifty White Tile Board - $19

After it has been sprayed with some whiteboard cleaner & conditioner, it will work very nicely as a whiteboard. Next, get some of these:

Dry Erase Markers, Assorted Colors, Ultra Fine Tip, 8 Count - $9.

By doing multiple plots and changing colors, you should be able to get a good idea of the variability of the sled’s movement.

Another (albeit more challenging) idea would be to raise and lower the marker with a computer-controlled solenoid.

[kyleschoen]

Working on this as well. Created a janky 1/4" adapter to hold a sharpie tip in the chuck and setting a pattern to “drill” 1/8" deep holes at 100mm spacing on the workpiece. Switched to 50mm at the edges since that’s where I think things skew a bit more. Then I’m going to tape up a large piece of paper and let the machine go. I’ll post the results here, but pretty simple and cheap setup.

I have some sharpie mini markers coming soon so I’ll redesign the adapter for those and might try to make a little spring setup so it doesn’t mash the nib too much.

[Dano]

I had the exact same thought as rich and the whiteboard . But I was concerned the dry erase might smear under the sled.

Floor covering paper and sharpie seem like a pretty inexpensive option.

The Home Depot 36 in. x 100 ft. Heavy Duty Floor Protection (2-Pack)...

Dano

[Betalombo]

@Betalombo

It’s true, I really want this. I would like to draw directly on the wood with some easy-to-clean material so as not to use paper. I was thinking about these 2mm [pencil] refills (Amazon.com)
Inside a 1/4 inch metal casing, driven by a soft spring that does not throw the graphite tip far when Z reaches the safe position, a spring that has a minimum travel of 5 mm

I found everything in my scrap: a 1/4" hollow pin, a medium-tension spring that fits freely inside the pin, which luckily gives me 10mm of travel. I pulled out all the disposable pens I could and started testing which tip would fit on my pin, and one fit so perfectly that it doesn’t come out and is completely centered. I enlarged the hole in this tip so the graphite refill fits precisely at 2mm, and I trim the graphite refills to 10mm outside the tip (thanks to the margin the spring gives me).
To define the Z home, in the case of the new refill, I lower the Z axis until it touches the material, then lower it another 7mm, then define the home and return Z to 15mm. To minimize the amount of exposed graphite and reduce the chances of breakage. When generating the G-code, I warn it that the safe Z distance is 12mm.
Please pardon my English, Google is helping me.
I cut a bathroom cabinet with my Maslow, the wood had to be cut on the back to make some pockets and I was surprised when I turned the parts over and saw a crack that crossed almost all the pieces on the front, but I decided to honor this mark of life of the wood, my wife gave me an image of birds from Pinterest and I redrawn it to make a DXF and my first drawing with Maslow,

the pencil test came out perfect, it doesn’t jump, it slides smoothly, it has a constant diameter, it allows me to use Z, because it is graphite I can erase it easily with an eraser, the ink does not dry out, these colored pencils also come in those diameters, when spring-loaded it passes soft obstacles, even if the graphite breaks it will continue to work thanks to the spring, but in the tests no spare part has broken

[Rich_Morin]

That link seems to be broken; try https://www.homedepot.com/p/The-Home-Depot-36-in-x-100-ft-Heavy-Duty-Floor-Protection-2-Pack-36X100HD2P/330278289

[boxomatic]

@Betalombo

I decided to honor this mark of life of the wood

Well done! That really turned out awesome, perfect “Happy Little Accident” turned into accent feature. Love you how continued it with the vertical along the side.

Is it a friction fit that keeps the graphite from falling out?

[Betalombo]

@boxomatic

Is it a friction fit that keeps the graphite from falling out?

Yes, the hole is precise, there must be friction, but the graphite is very smooth, the spring must compensate for this to push but it must not be too strong so as not to throw the piece when moving Z.

[TimS]

Curious on the “smudging” factor with dry erase. Excited to see test results. Love the ideas of a reusable board and multiple colors.

[kyleschoen]

Seemed to work very well. Going to measure the results and post them to my other thread about the X and Y axis being off by 1%. I’ll cross post here too.

Initial results: looks like the X axis is pretty accurate until about 2" from the edge, then it’s short. the Y axis is similar but long near the edges. both are about 2-3mm off, and worse in the corners. Also noticed that there’s a shift depending on the direction the machine is moving in the X axis, by maybe 3-4mm which is odd.

Worth noting, this is a vertical frame, I extended the anchors by 1’ in the X axis and my arm order is different, and I used manual measurements instead of the machine’s calibration since another user said that was more accurate.

[dlang]

Kyle wrote:

Initial results: looks like the X axis is pretty accurate until about 2" from the edge, then it’s short. the Y axis is similar but long near the edges. both are about 2-3mm off, and worse in the corners. Also noticed that there’s a shift depending on the direction the machine is moving in the X axis, by maybe 3-4mm which is odd.

please check the frame size calculators:

https://lang.hm/maslow/maslow4_frame.html

Anchor Point Layout Simulator

is your workpiece completely within the green? or are you hitting the edge of
the green when you start running into errors?

These calculators are based on theory and calculations, we don’t have enough
feedback to know if they really show the problem areas or are overly
conservative or overly optimistic

David Lang