I can only do that with trochoidal and even then the plate gets incredibly hot in the end. Everything else needs cooling as far as I know. Or you need to find the perfect settings… 
Are you trying to cut aluminum with 3D printed XZ plates and without strut plates?
I’ve have had my LR3 for 2 years and I still haven’t cut aluminum.
Just saying…
Yep! I gotta piece strut plates together because my Y-axis is shortened. I guess I got ahead of myself with cutting the XZ plates before strut plates. I figure nothing beats a failure but a try? 
2 Likes
You are adventurous. I’ll give you that. 
1 Like
We are about to get some settings so we can add this thread to the Al tag and get more people cutting their own with printed plates!!!
2 Likes
The good: (pics and video below)
It cut great with aluminum this time! I used the following settings, following the recommendations @vicious1 made up-thread:
1/8” V1E.com endmill
200 mm/min feed and plunge
0.25 mm step over
0.5 mm step down
no coolant
same dull-ish endmill as before (I just wanted to see how it would do and if it was totally ruined, it still felt pretty sharp)
The bad:
It looked great coming out of the board, but I think my work-holding arrangement has some play (superglue and painter’s tape) which led to some parallelogram results as Z went down, and also I think I probably need to re-figure my steps-mm because the part isn’t accurate. It should be a perfect square 1” or 25.4 mm in width and length, but instead it’s ~24.5 mm in the X-axis and ~25 mm in the Y-axis.
It cleared chips like a champ, never got bogged down. There was probably some deflection and chatter, especially when starting out a straight plunge cut for the 1/8” holes on the test 1” square I was carving.
Time-lapse for fun: https://photos.app.goo.gl/Hw2NNMTDVB9Re1GT7
The finished product:
2 Likes
awesome!
did you do a full depth finishing pass? 0.3mm is a good starting point
I’m not sure! 
I can’t find any settings for a finishing pass in KiriMoto on the job settings (left side in current version, stuff like defining stock, limits, G-code conventions like ease down) or in the cutting operations menu (right side in current version, for me Pocket to drill holes and then Outline to cut the inside and outside square edges).
What does a finishing pass do? Do you think that has something to do with my parallelogram’ed results, my weird dimensions, or both?
Oh @vicious1 I see what you mean I think after reading some of the KiriMoto docs. You might use Rough to remove the bulk of the material between the stock and the part but leave a margin closer to the part untouched and then Contour to do 0.3 mm step down finishing with more detail (and removing less material) for the closer operation to the finished part? I simply used Outline in a single go (which it seems like can be used for both, but I certainly wasn’t careful with it).
1 Like
This might help.
2 Likes
OK I figured out the parallelogram and bad dimensions last night, I had slack in my belts, mostly in Y-min but really all of them had slack. I think that explains the weird dimensions, I just fixed it and once my wife wakes up and kids are out of the house I’m going to repeat that test cut and see if I get more consistency.
EDIT: I just realized I may have to re-do steps/mm now that I’ve taken the slack out. It probably wasn’t accurate before.
1 Like
Progress still, but not quite there. I’m seeing about 0.4 mm of error after re-calibrating steps/mm for both X and Y following tightening the belts. The part did come out pretty square though there’s a tiny bit of skew and play I can pick up with my most-accurate (non-engineering) square. I hadn’t re-done my skew fixes from before, just steps/mm, I’ll do that before the next test-cut.
Do you all think I just need to continue to dial in steps/mm? Maybe I need a more accurate measurement from dot-to-dot, I’m using a metal carpenter’s yard stick. Perhaps there’s some error in there I’m translating to the machine.
1 Like
Nice test cuts!
- Adjusting Belt Tension, then, calibrating steps/mm by using largest motion possible that you can accurately measure (e.g. moving 1000mm along Y axis) seems like a good sequence that you already did.
- Already checked your end mill dimension matches what the CAM software has configured (KiriMoto or Estlcam) ? Am asking incase you have 1/8" endmill but CAM mistakenly has 3mm configured, something like that could contribute.
- Already doing a 5% finishing pass to help part edges be more dimensionally accurate (and smoother) ?
- Endmill is in good shape and not chattering or deflecting?
- Consider cutting cheap ply/mdf 100mm square if trying to verify CAM, endmill and machine are tuned as much as you’d like. Cutting larger cheaper material might help reveal what to tweak next.
Sub mm accuracy is good, and no one expects micron accuracy, but… I don’t know what’s acceptable/normal precision and accuracy for a given span/dimension. Partly because depends on so many variables, rigid struts, rigid XZ plates, rigid pipes, nicely snug up bolts, no Core-to-pipe slop, no excess Core-to-pipe friction, printed part material/infill/walls/adhesion/layer-height and other stuff probably.
Cheers for sharing your Alu adventures!
1 Like
I’ll double check the tool. I haven’t really done any sort of finishing pass, I’ll need to check up on that in KiriMoto. @vicious1 mentioned something like that last night as well. Maybe the Outline operation is simply too rough.
1 Like
You absolutely need a finishing pass for accurate parts.
1 Like
Wait… you got these cuts with a (mdf/ply/hardboard?) strut, plastic printed XZ plates and no finishing pass? If so, that’s pretty cool imo.
What usable cutting area, 4’x8’ or 2’x4’ ?
1 Like
Well the hard part here is you are trying to calibrate you machine with your first aluminum cuts.
The side plates are not that critical, cut them, install them, recalibrate and lets do wood test cuts first. We are kinda in the realm of chasing zeros here, way to many variables, starting with printed XZ plates.
1 Like
You know, I think you’re all right, but also I think @azab2c nailed it with checking the tool. For some reason I had 3mm flute instead of 3.175 for the 1/8” endmills I got at the shop. I think that’s probably introducing the error on every side, leading to the .4 mm error.
I’ll try one more cut with the correction, and maybe also try to nail down a finishing pass if I can figure out how to do it. That should put me in the ballpark for pretty darned accurate, and then I’ll cut the plates. Woohoo!
@azab2c I’m using your 3D printed XZ plates and not even wood struts, I still have the temp placeholder struts in. And yeah…trying to cut very accurate aluminum parts for like…my first thing I’ve ever done on a CNC. I think @jeyeager thinks I’m nuts 
. Usable cutting area is like 36” x 15” ish.
@vicious1 I think you’ve nailed it, I’m probably majoring in the minors. Once I correct the tool definition and maybe try to figure out a finishing pass thing I’ll do one more cut then I’m gonna get these things done and installed. See you guys on the flip side at which point I should have metal XZ plates at either end and proper dust collection, then I’ll cut MDF struts and rewire everything final.
5 Likes
Cutting 6061 Alu with just temporary plastic struts. 
You are bold. Not recommended, and strays from yellow brick road path, and docs/forum that recommend cutting and installing dimensionally accurate final Strut, before taking on tougher materials. Seems like you know this already, and are having fun experimenting and trying stuff out.
That said… Your time, material and your machine, and, being honest… am curious to see what happens 
2 Likes
LMAO, OK I’ve got feeds/speeds/etc. down. I figured out how to do a full-depth finishing pass in KiriMoto even! And now that I’ve a) set up the finishing pass to increase accuracy and b) adjust the cutting edge of the tool from 3mm to 3.175 mm (1/8”)…the error has now moved from being .4 mm under in X and Y to ~.45 mm over in each axis. For today…I give up .
3 Likes