I have a 3 XXL. I currently got a DeWalt router on my unit with a 2.2kw and VFD waiting for parts to complete it’s installation. But I digress.
I have a project I’d like to run, but it’s run time is pushing 24 hrs. Doesn’t give me a warm and fuzzy feeling.
So I’d like to know: What’s the longest program you’ve done?
I have ran several 14 hour projects. During the run I would feel the router and the stepper motors and they were warm but not hot. I also ran my Jet DC1100 dust collector the whole time and it was cool as a cucumber. So I think a 24 hour run would work. The issue is staying awake. I would suggest you break it up into smaller chunks so you can stay and monitor the cut. Leaving a Shapeoko running without supervision can be dangerous.
With dust collection and the fan on my Dewalt 611 router it has never felt hot or too warm to touch. So between the fan in the router and the dust collection I have a lot of air cooling the router. If you are using a shop vac that could be the X factor. Shop vacs are not meant to run continuously for 24 hours.
Well, you didn’t ask, but I would suggest you present your project to the forum for evaluation of your design parameters. We’ve seen long runtimes before, and easily reduced those with suggestions that you could evaluate. On another forum, I’ve seen 12 hour cuts reduced to 2 hours.
Unfortunately I have to rely on an online STL to gcode converter. So I don’t have very many alterable parameters to trim time off. If it starts to cut well I can do a feed override in carbide motion to trim more time off. But I won’t know that until I’m already in it. I have yet to decide on a software suite, until then I don’t have that many options available. The two parameters I did select used a 1/4” flat end mill for the roughing pass and then a 1/8” ball nose finishing pass, The latter making up the vast majority of the work time.
Wow. I never would have guessed that. Does the converter ask anything about your machine before creating the gcode?
@fenrus has a tool which will optimize the feeds and speeds in a cut — might be it would help.
Most likely you could edit the G-code file and split it up into sections and send them.
That said, I’d recommend a test which uses either a partial section, or a scaled-down version of the design to evaluate things and determine how the estimated cut time relates to the actual machine time.
I would not want to do a 24 hour run, i would figure out how to break it up.
I wish i could just load the program and let it run, but things happen, and if something happens it’s usually not good and i’ve had to stop it to make sure things didn’t get damaged.
I’ve had the spindle ram into clamps after a bit change, because it didn’t start from the original position and then a clamp was magically in the way that wasn’t before.
I’ve had it get stuck on a plunge and then the x/y axis slips and then everything is screwed after that.
And the list goes on…
I never run my machine without at least being in the shop, you can usually hear when things go bad.
if it ends up being the one I wrote – there’s a bunch of semi hidden parameters that can be tuned…
in theory it is possible to also do a 2 phase approach … think “all even lines first then all odd ones” so you can restart but … slightly messy
Run your roughing pass and shut down. Come back next day to do finishing pass. X y and z are remembered.
Yes. It puts out a gcode that goes directly into carbide motion.
That is preferable. Need to figure out how to do that. May have to select the software I want to settle down with sooner than I was planning.
Definitely. Not sure how to accomplish this with my current set-up. Looking into more advanced software to be able to do that.
It is tautological, but you may not know what you don’t know so be careful applying that gcode.
You didn’t say which online gcode convertor you were planning to use, but the of the few I looked at, they were for 3d printing, with the exception of one that didn’t state what type of machine the gcode was for.
It would be well worth your time to at least look at MeshCAM to generate the gcode.
Pause it every 4 to 8 hours for a break then restart.
That link goes nowhere. Either way, I still feel it’s worth while to pay for software if it works for you, and it saves time and headaches.
I totally agree. Paying for the software with pay big dividends for this project and I’m sure many other in the future. It sounds like you will go that route sooner or later, so might as well get the benefit now.
But to answer the question, the longest project I’ve done took 12 hours. I broke up the job in 3rds using the tiling option. I’ve also done a couple of 7-8 hour jobs. Way too long in my opinion.
I use CC and have made inlays using the Vetric method. That method is similar to what CC does but for the male plug you have a starting depth of .1" That effectively makes the male inlay slightly smaller so it fits into the female pocket better. The down side is you have a big cavern under the inlay. So far I have given up on the CC inlay because I have not successfully gotten a good male inlay fitting into the female pocket with no gaps. I have sent my project to projects@carbide3d.com and got a few spotty answers but no resolution. I like wood inlays but usually just do an advanced vcarve and fill with epoxy. Carbide Create has come a long way in the last few years but still has a way to go with the inlay mode in my humble opinion.
You can even mill cut threads in Vectric Vcarve. Thread milling cutter not included.