Any ideas how much desire NOT to sand I can cover with a surface finish of sorts? Can I get to maybe 320 grit and finish with a poly and get that glass look back?
There are a variety of options. Arguably the best one if your project lends itself to it, is…pouring a final thin layer of clear epoxy onto the piece. It’s like magic, the original machined surface won’t show and you end up with a clear shiny finish. But of course, many times you can’t do this (e.g. if you would like to keep the wood surface alone)
What I often do is sand it manually, from 320 down to 3000 grit or so. But I’m really no expert, and I seem to remember folks mentioning a number of good tips, let me search…
Nice, thanks! I will definitely see if I can pull off leaving the shiny finish. I think I can actually, I’ve left enough depth in the first contour (4mm) such that I should hopefullybe able to full up with maybe 2mm to spare, and then I can face the wood up to 2mm without touching the epoxy, but removing the seal coat epoxy from the surface of the oak.
That said, next time, I’m going to look at wood that’s a bit less … fibrous ? then oak? Or maybe run faster, spring pass, etc etc. A few curlies I had to try to clean up then said forget it.
I feel down the STL rabbit hole today with a short piece of 2x6 in hand.
I grabbed the free STL demo at 3dwave and loaded it into the STL to PNG tool@fenrus wrote. BTW, thank you for making that available.
I imported the PNG, made one roughing and two finishing tool paths and presto, I had a pretty nice looking chuck of pine 2x6 in hand. I brushed a little oak polyshade over it for some highlights.
@greg5 Actually, you do have a planer albeit, one with a very narrow blade, your Shapeoko! Smoothing the top or last layer pour works great and your don’t end up with plastic dust from sanding everywhere, just chips
For my final finish, I’ve had great success spraying with a polycoat type finish, satin, gloss or somewhere between after sanding with 220 when finished smoothing. This thread dives deep into the options and finishing techniques by several users including @AndyC and @fenrus , and others, as @Julien had mentioned. Looking forward to seeing your final result!
Thank you. It’s posted on CutRocket.com. Not sure if they have it available or not. Haven’t been there in a minute.
Went and looked and it is posted under “Household”.
Got my feet wet with my new Pro XXL today. Found the “Friends Frame” on CutRocket and decided it would be a good practice run since there was a bit change involved.
Itty bitty snail, testing out a ball socket for joining jewelry pieces to create movement. Think the antenna is pushing the wood a little too far at this size though
It’s a very deep rabbit hole, but I’ve got it fully up and working. The first hurdle being that the standard electronics only supports XYYZ axes, with no spare capacity or IO pins on the controller. I researched various alternative controllers, but settled on the KeyeStudio MKS 1.2 combined Arduino 2560 and RAMPS 1.6 on a single board, hosting the Grbl Mega 5X branch of code.
Specs:
KeyeStudio MKS 1.2 Arduino/RAMPS Controller board, 6 channels plus the usual limit sensors, PWM, probe IO etc
Step-Stick 2209 drivers x 5
Power supply 24V 15A with fan cooling
Grbl Mega 5X branch of Grbl uploaded using Arduino IDE
“from Amazon” 4-jaw spindle with Nema 17 driver and 4:1 belt drive gearing
Aluminium case for the electronics with miniature cooling fans
Universal G-Code Sender (UGS) with ABC axis enhancements (hopefully due to make 2.09 release)
Existing Huanyang VFD with 800W water-cooled spindle
Fan water reservoir/cooler with modification to engage the fans only at 28c
Laser cross-hair XY zero ‘pen’ “from Amazon” enabled from a spare IO on the controller
Video of the axis running, making a test piece:
The electronics need to be tidied up… but hey, it’s all working
Wow very nice. I can’t help but wonder if there is interest in a custom controller board that fits in the existing enclosure but adds an extra stepper as well as a separate PWM output for laser, and anything else that can fit.
@ctdodge About $250 all in, plus self-time putting it together and configuring Mega 5X, and contributing to the UGS code development. Will itemise the costs when I get a moment