We use cookies to personalize content, interact with our analytics companies, advertising networks and cooperatives, and demographic companies, provide social media features, and to analyze our traffic. Our social media, advertising and analytics partners may combine it with other information that you’ve provided to them or that they’ve collected from your use of their services. Learn more.
This is an interesting approach. Your test pockets have plenty of width, good for chip clearance. One thing you might wanna try is avoid going into very tight corners. In order to take this (limiting) parameter out of your test, I’d say try a corner radius of 2x your bit dia (either by adding fillets in CAD or by telling Fusion what your min radius should be in CAM).
Another approach would be to find a sweet spot for your tool’s chipload ( => chip thickness) starting at low speed & low feed and then progressively go up. One step by proportionally increasing both (same chipload, higher MRR), one step by increasing feed rate only (higher chipload, higher MRR).
John Saunders of NYCCNC has some nice videos demonstrating this. See here:
and here:
Two notes on John’s approach:
he uses a mill, he can start at low rpms (a router cannot, you need fewer flutes instead)
he ignores chip thickness, he uses chipload (actually he was asking in the comments how to calculate it - he is lucky because using high DoC his chip thickness is not that lower than the chipload)
First test complete bit no time to analyse tonight. My rig managed to complete the full carve with no lost steps or any malfunction in that regard. The gantry is showing its weakness (3/4" MDF) during the more aggressive cuts with skewed sidewall of the pockets.
This image show with highlights where the sides started to become skewed, highlighted in red.
The rails the gantry rest/move on is also oriented in a way that allows the side plates to “rotate”
For CNC rev.3 this will be improved among other things I have learned since the beginning.
However, I ran it quite hard during this test, WoC=0,8mm @ 600mm/min so I might dial WoC down a little bit and see how it may act differently. Running a similar test again, but with a fixed pocket depth and variable WoC might provide me with even more data
Due to rough running/chatter I increased RPM to about 13-14k which smoothed things out a little bit.
BTW - the complete carve, 8 pockets, took 12minutes
Nice big chips! The alu. block was luke warm to the touch after completed cycle.
Denatured alcohol used btw with a mister.
Adaptive Clearing is just a roughing pass, for production runs a finishing pass following a different strategy will also be required.
@C.j.Shull This is true. To give the complete picture, depth of field is affected by aperture, distance, focal length and medium size.
the smaller the aperture (higher F stop)
the longer the distance
the shorter the focal length
the smaller the medium (sensor)
the higher the DoF
In the example above aperture alone would probably not be capable to provide sufficient DoF, as the distance is short and the focal length is medium to long.
An alternative could be combining multiple shots in post processing (focus stacking).
@AngusMcleod Really? Nikon in the Learjet? I’d expect a Phase One to say the least.
You don’t need a tilt-shift for this. I use a 50 year old manual focus Micro Nikkor 55mm with its 1:1 extension tube on my D750 with no issues at all. You could probably find on in good enough shape for under $100
No worries…
I have a complete Hensel studio lighting, Canon 1DsII but very little time to play with them - hence the crappy snapshots
Relative high ISO, low aperature and handheld, under CFL (poor) lighting.
Havent had time to perform further testing / system optimization as I need to do a couple of paid projects that is lined up
After not being able to carve much for a while I finally had some time to test some changed GRBL-settings, specifically acceleration. Also shown is the mister.
Pocket 16x16x4mm drawn in Easel, nothing fancy.
Max move rate (G0)
XY=5000mm/min
Z = 3000mm/min
Accelleration (GRBL) for XY = 500 and for Z = 750
I really haven’t played with the plunge numbers except keeping them on the low side.
Tonite had a first for me, a completed 3D carve of aluminium based on a quick arbitrary shape in F360 with a 2 stage carve process. Just a practice run to get properly aquianted with the whole process.
With the current speeds and acceleration settings there is not a whole lot of time to react if something isn’t right lol!
Grabbed a F16 Falcon image of the net and ran a quick image trace in Easel, tweaked it to my liking and hit Carve.
Came out decent, the floor of the pockets are a little ragged due to step over is to high / pointy bit.
After hitting the HDU with satin black paint I wet sanded of the excess. Result was quite pleasing
By eye, I made 4 tiles (rectangles, set to cut on outside of path) and placed them approximately where I wanted them. Then edited their final position so they all had the same distance between each other.
Looks good up on the wall!
Very happy with the end result, this could be one area where I could get real creative with airbrush and different colours, monochrome or shades/different colours.
I fly gliders (sailplanes) and one of our planes in the club fleet lost its small canopy window vent. Decided to fix this using my CNC and quickly made up a sample part with Easel. Since Easel only will do flat surfaces I also modelled the required curvature in Fusion360 and made a tool path for detail pass. Two passes, 45deg diagonally and +90deg for perpendicular pass. Buffed the carved surface back to shine.
Never worked seriously with Plexi before but it worked like a charm. Glued the piece to its angled bracket using Acetone.
Pictures tell 1000´s worth, so here is a few:
Test piece to verify dimensions and radiuses (55x55mm)
5min carve with Easel (Feed 1200mm/min, DOC 1mm with 3mm 1F end mill, 40% stepover)
No worries…
