Trick Laser MAX METAL with Linear Rails

[mhackney]

I picked up one of the MAX METAL sets that Eagle sold last month and I'm starting the build now. I basically want to do all the bells and whistles on this machine. Here's what I have in mind (and in boxes ready to assemble!):

• MAX METAL frame
• .9° steppers
• AstroSyn vibration dampers
• LM 76 Speed Demon SG type size 10 linear rails and slides
• home brewed CNC milled aluminum effector
• home brewed CNC milled aluminum carriages
• home brewed mag ball delta arms - thinking about using a very stiff aluminum tube
• optical end stops
• FSR probing
• 24V power supply
• Duet electronics with PanelDue panel and WIFI connectivity
• silicon 24V heater
• 1/4" MIC-6 cast aluminum heat dissipator and bed with PEI attached direct - no borosilicate glass
• hot end TBD but I'd like 2 or 3 filament switching extruder
• as many BondTech QR extruders as needed by hot end

[mhackney]

I started assembling the frame - dry run - so I can figure out the linear guide install. Since they sit further out from the uprights, I'll likely have to add additional idlers to position the belt properly so it is parallel to movement. I did run into one gotcha - I ordered 700mm rails thinking I would cut them to fit. I did not realize that they have steel rails inside the channel. I need to call the manufacturer to see how best to cut them as they interfere with the cross pieces now - as shown below.

.9° stepper on AstroSyn damper test fit

Speed Demon linear rail - just a bit too long. The holes are M3. I have some M4 threaded quick insert T-Nuts that would work perfectly on this uprights that I'd like to use. I only have button head M4 screws and they are too large for the countersink. Mot sure about regular M4 cap screws. I need to order some and see. I might have to bore and counterbore the mounting holes. But the rails themselves are a good match to the 1" 80/20 uprights.

Where the rail extend up past the bottom of the cross pieces it won't sit flush to the upright since it is a little wider than 1". So I do need to figure out how to cut the rails to size.

[mhackney]

I just discovered that LM76 (linear rails) are manufactured in East Longmeadow, MA not too far from me. That's why they were drop shipped from there. I might be able to drop them off to have them recut.

[JFettig]

how do you plan to mount them without interfering with the belt path?

[mhackney]

Like I said, I will likely have to add additional idlers to move the belt towards the center of the machine and to clear the linear rail. Not sure if the V2s have this but the original V1 has 2 idlers at the stepper that position the belt proper in the track. The other option would be to do a Kossel like stepper mount and belt but I'd rather avoid that.

[Eaglezsoar]

Thanks for posting this Michael. I love to read anything you post because you have great pictures, a technical but friendly text and your posts are just great!

[ramai]

I'm the other guy who bought eaglezsoar's kit. I'm definitely going to follow along here. I've been gathering all of the pieces together for my build, I can't wait to get going on it, and your posts here will give me inspiration.

[mhackney]

I just talked to a tech at LM76. Other than the fact that the rails are RC60, they can be cut to length. I'll probably do it on my bandsaw with a carbide blade slowed way down and mist coolant temporarily rigged up. I only need to remove 10cm or so but I'll wait until I sort out other aspects of the build before cutting.

[ramai]

Are they hardened steel or something? Seems like you could just pop them in a square box and hack saw the tip off.

[mhackney]

Yes, RC60 is the hardening designation. It would be tough work with a hacksaw but doable with enough time and perseverance (and good blades).

[ramai]

How come you are choosing Duet over smoothie? After researching myself, I decided smoothie was the way to go, but maybe you can convince me otherwise.

[mhackney]

I am not here to convince anyone what to do. I ran smoothie on 4 delta printers and even contributed the TemperatureSwitch module. I like smoothie, my only "issue" is lack of delta calibration that fits my work flow. The Duet dc42 firmware has several advantages in my opinion (and I've experienced first hand now):

1) its delta motion control calculates EVERY point of movement, not short line segments like all the others do
2) its auto calibration takes less than 30 seconds to run and works every time on all of my deltas
3) its built-in web interface is really well done
4) the PanelDue display is big, full color and touch screen, and works great
5) the Duet board itself has more memory and a faster processor - memory being a critical issue with current smoothieboards

I am running smoothie on one of my deltas just to keep current on smoothie development. I don't know who the winners are going to be but these are two major contenders. I do have one other slight bias - I think the Smoothieboard/Smoothieware approach of being a general purpose machine controller makes it more complex whereas Duet/RepRapFirmware is focused on 3D printing.

[ramai]

Exactly what I was looking for. Even though you're not here to convince anyone of anything, you have a lot of experience, and I find your opinions on these sorts of things quite persuasive.
That's interesting about it calculating every movement. I wonder if that means other firmwares are reducing the resolution of the gcode.
Thank you for your input!

[mhackney]

Other firmwares are interpolating movements into short line segments so, in theory, it has an affect on print quality. There are other factors at play too though - hence the reason for "in theory". By it's very nature, 3D printing starting with an STL file is going to sacrifice some resolution depending on how the CAD model was converted into an STL (most profound is the # of triangles in the final model). Here is an example of how this works...

Imagine you have a CAD drawing of a 20mm diameter sphere. The CAD tool uses sophisticated math to calculate the geometry and display it. It looks like a perfect sphere on screen. Now, let's run that perfect sphere through the meshing process to create an STL version. To demonstrate I am using 5 different resolutions that generate a low number of triangles to a high number of triangles. When you download an STL from Thingiverse, you have no control over this, the original author made the resolution choice for you. And frankly, most of the folks uploading to the shared services really don't know what they are doing. Anyway, back to the example - here I've meshed to create a "sphere" with 48, 224, 960, 16128 and 99856 triangles (from left to right). You can see that the far left sphere is course and the one on the far right is reasonably smooth.

Now to take this one step further, imagine your slicer creating gcode from these STL models. I created an imaginary slicing plane at 12mm above the base that is .02mm thick and took a cross section of the model at that plane. This is the red line you see inside each model. I projected these slices up above the models so you can see them clearer (black). The 48 triangle model has a slice with 8 sides - pretty low resolution. You can see as you move left to right, each slice has more line segments. These will result in a finer/higher resolution print all things being equal. But they are not! In addition to the slicer converting curves into short line segments, the firmware also imposes its own line segment on top of that - that is all of them EXCEPT dc42 RepRapFirmware. dc42 draws each point along the way completely tracing the original path it was given. Smoothie, Repetier, Marlin and the others actually break the path up into short line segments and draw those. This does not always match the original path exactly.

As you can see, if the STL had a low triangle count, you are going to get a course print. Increasing the triangle count will improve the print quality until you reach the segmentation threshold of the firmware, at which point, increased # of triangles won't have an effect and might actually make things worse. That is except dc42, which will faithfully trace each of the tiny line segments exactly point by point.

[critical_limit]

... Increasing the triangle count will improve the print quality until you reach the segmentation threshold of the firmware, at which point, increased # of triangles won't have an effect and might actually make things worse. That is except dc42, which will faithfully trace each of the tiny line segments exactly point by point.

That´s exactly the reason why I changed my smoothie to the duet. DC did a really good job and I love his firmware - and the web interface.
With my actual homebrew Kossel I´m on 0.9° steppers and small pulleys so I have 200steps/mm. And if I print my own made parts with much triangles as I can get out of the design software, you can see it in the print. The duet print results look better. Specially on high resolution prints with 0.25 Nozzle. For me there is no reason to go back to segmented movement - exept maybe the 626pilots routine for bed mapping which is (actually) missing in the DC´s fork. Actually I´m printing with a 0,15mm first Layer and that is working fine. My PEI Plate is within 0,08mm (after much of sanding). With a bed map and, let´s say 100 probing points this would be amazing and would bring me to a first Layer of < 0,1mm. But in the moment I´m absolutly happy with DC´s firmware and the results I can get.

[mhackney]

I posted this link here last week: http://v5.rhino3d.com/video/01-what-will-3d-print

It has some great information on what makes a good STL and how to create one (using RhinoCAD but the ideas are applicable to anything).

[626Pilot]

1) its delta motion control calculates EVERY point of movement, not short line segments like all the others do

I was actually thinking about porting the Duet-dc42 positioning algorithm over to Smoothie, but after I compared the two parts side by side, I didn't think it was worth the effort. This image describes Bresenham's line drawing algorithm, which is used to render diagonal lines on displays that can only perfectly produce orthogonal (perfectly right-left or up-down) lines:

Bresenham.png (4.68 KiB) Viewed 20834 times

3D printers have exactly the same problem. The steppers can only be at discrete positions, so any improvement in position generation beyond that point gets eaten. Similarly, the filament diameter never stays exactly the same, and there is also die swell to consider.

I've held a part printed with my Smoothieboard next to a part printed with my Duet. There is zero visible difference. Smoothie is generating segments 200x/sec. Even if I printed at 100mm/sec (and acceleration limits let it get up to that speed), it would still be generating segs once every 500 microns, and those will have only a small amount of error. Judging by the identical print quality, I would say that error is either being hidden by stepper resolution (which is 0.9 degrees on my printer), smoothed over by the molten plastic, or that it's so tiny as to be invisible to my eyes. The Duet produced no fewer Bresenham artifacts than the Smoothieboard. Top surfaces still have the same familiar "slightly bumpy" feel.

If someone can show me a side-by-side photo of something printed with Duet-dc42, and then the exact same object printed with Smoothie, and there is a genuine difference that I don't need a 400 power microscope to see, I will reconsider porting dc42's method over to Smoothie.

Also, the Smoothieboard is over 30MHz faster than the Duet. The Duet still wins on RAM (96K vs 64K). I'm currently working on my Smoothie fork, and will be reducing RAM usage if I can.

The two things I miss the most about the Duet are the web interface, and the PanelDue. Someone is working on getting Smoothie to work with the PanelDue. The web interface, however, is awfully disappointing compared to the Duet. It seems like someone hacked it together in a weekend and hasn't touched it since. That will take longer to fix, and I probably won't be the one to do it.

[Glacian22]

If someone can show me a side-by-side photo of something printed with Duet-dc42, and then the exact same object printed with Smoothie, and there is a genuine difference that I don't need a 400 power microscope to see, I will reconsider porting dc42's method over to Smoothie.

This isn't a side-by-side, but I've never seen a vertical surface printed on a delta that was completely free of banding like this: http://forum.seemecnc.com/viewtopic.php ... 175#p78990. Do smoothieboard prints look that nice? That's a genuine question, I don't have one and I also haven't seen any relevant images.

[626Pilot]

Here are two pieces, very similar geometry, same spool, one printed with Smoothie and the other with Duet-dc42. They are spool holders for the MAX METAL. If I didn't already know, I wouldn't be able to tell which was printed on which.

N.b.: The one with lots of stringing was when I was using fast retraction speeds. The one with no stringing is after I switched to a retraction speed of 15mm/sec. It really works!!!

[ccavanaugh]

626Pilot,

An effort to port the Duet motion planner may fix/uncover the bug with Smoothie firmware not being able to handle the high resolution files that Simplify3D sometimes creates. I would be willing to contribute to a bug bounty if you ported.

It could also be a win for those of us with 1.8 degree steppers. My next Max Metal build will get 0.9 degree steppers, but I'm on the fence between another Smoothie or trying out a Duet.

[626Pilot]

I don't know what bug that is. You can tell me about it in my thread (see sig).

[mhackney]

Ok, I have the belt path issues all sorted out. I made new idler mounts for the top and a special one for the bottom.

First, here is the belt mount to the top of the block. I am making a belt tensioner for the bottom side.

Top now has 2 sets of bearings to offset the belt properly.

A better view of the belt path.

This is the bottom mount to offset the belt. There is not a lot of room to work but this works. There will be two M3 screws holding the bucket onto the rail. As you can see on the back side of the mount laying on the cross rail their is a recess milled into the mount to help align and brace it. The bearings are perfectly centered and there is absolutely no flex in the mount.

I am going to have to open the width of the slot a little bit so the bearings clear. I don't want to take the entire frame apart for this so I'll probably use my Foredom with a small grinding bit to remove the aluminum.

[bot]

I'm excited to see your new printer come together! Those rails do look interesting. The specs on their site seem great. Not sure if they're listed in metric or 'merican, tho. Do you know which it is?

[mhackney]

Metric specs for dimensions. The rails I got are just a smudge wider than the 1" extrusions for the uprights. Now that I have my frame together, I can measure the actual length for folks to order. I got mine too long and had to bevel the edges a few inches to get them to fit.

[Xenocrates]

Now here's a question Mike, Just how expensive are these rails? I looked at one of the distributors which have prices online, and it's 47$ a carriage, and 98$ for a 4 foot length. Did you order from them directly? If so, are their direct prices more reasonable? Will they give you a quote by email or over the phone? What sort of lead time do they have?