Done and it begins,,,,,,hmmmmmmmm

Well the machine is done.

I wish i could pour out amazing enthusiasm good wibes, but hmm not really.

The build went ok, instructions left a lot to be desired, I consider myself fairly experienced with technical and electronical buids and i found the instructions confusing at times, mostly because small details are left out that later proves to be critical, also the photos often stray from the build sequence described.

At the final staged it was very annoying to find that inventables apparently aimed for including parts down the exact number of individual nuts and bolt, of course i was missing several. Of course I may have messed up, but I find it hard to belive that many went missing when I make a point of doing the build in a dedicated area and throw nothing out before I am done, anyway I would really expect a little excess of these tiny generic parts in a kit of this price range.

But using parts I had I was able to get it all togheter.

First run, all the excentric nuts came undone and machine failed completly, this design it terrible, at the very least they should use some kind of nyloc or locking nut, well i added some and the machine ran seneral stresstest fine, drawing more or less perfect circles in mdf.

So i went for a proper project and mounted up a piece of acasia wood, not to long itno the carve its apparently loosing steps, stopp lover feedrate( why this cannot be done realtime is beyond me) now it looks ok, after an hour of monitoring I leave it to it. Come back later and the whole board is torn to shitt like it had mental breakdown, and this really brings us to the next flaw. This machine is doomed to be loosing steps by a huge number of possible casues, why not build in some kind of tracking device or some electronic function to meassure load on the engines so it could abort when it was loosing steps? its completly beyond me and in my oppinion the only way you can get some reliable function of a machine that relies on belts.

Well back to start and try again, very low feedrate this time, an hour into it it looks very controlled and fine, so I leave it to it, come back to find the power went out, my fault for plugin in the EV in the garage on the same circuit. The while trying to pick up the carve again I set up the home point again to be sure(made a mark the first time) and prss the button “raize the bit” only to my amazement se that the machine does the opposite, it drives the bit full force into the board and litteraly tears itself apart, the is no abort button, no stopp no nothing so only stepping back in the browser finaly stopped it and let me raize the bit again. So now the gantry a possibly a whole load of stuff is bendt to hell, the gantry is 3 mm higher in the center now than on the sides, god damn piece of shit software, and why does the software alov you to smash your stop switches when jogging, what could possibly be the point of this?

So all in all I was recommended this machine as easy for beginners and that it should be able to do light production, I would say absolutly not, this is a machine for people who wants to play around with a cnc machine and take joy in the challenges it produces and improvements they can make.

Not really sure now if this thing is worth my time and now I have no clue as to how much parts i need to change to get it back into shape.

I feel your pain and I think that must of us felt the same way at the beginning of our purchase of the Xcarve. One thing is for sure, you will have to be on top of this machine in order for it to do descent cuts. Is not going to stop, period! I guess, if you have he patience, you might get used to it and with the help of everyone here on the forum, you will be able to solve a lot of the issues. I don’t cut as much because of the frustration sometimes. I decided to stay with mine and not sell it because sometimes is a joy to cut things with it, but yes, I wish it was more reliable but I guess thats why it was cheaper than others. I hope that you can spent more time with it and get most of the issues worked out. I bought mine like 8 months ago and is still not 100% dialed in, ( I broke a limit switch and I left it like that, those things are a joke). But it cuts pretty descent and Im ok with it. Try to give it some time and play with it, you might enjoy it once you have it somewhat dialed in. Good luck!

The issue is really what it needs to stay in shape, not the number of issues to dial in, I dont really see much potential of this thing ever running reasonably realiably to the degree that i would dare mount parts or materials of any value in it, and then its of no use to me really. in my oppinion it would be so easy to remedy the most basic flaw in this machine. If it could abort once steps was being lost , then it would be fine, I could very well deal with that and some sort o realiability could be had. This could be a very simple sensor measuring movement betwen the acess, optical, mechanical, electrostatic,magnetic, laser, infrared, take your pick anything would add to this.

I’m not trying to be smart here, s please don’t take it that way.

That’s the beauty of open hardware/open source. All of the hardware files and software is available for download. You can modify your system any way you see fit. If this doesn’t get officially added, implement it. I think we’d all find something like you described interesting.

The software/firmware side is where it would get tricky, but I’m sure it could be done.

I understand the frustration, when I put together my kit is was the Shapeoko2 and I was so frustrated by the unreliability of the machine. I ruined so many pieces of wood. I would start a cut and since I was using the very underpowered 300 watt quiet spindle my feed rates were very low so everything would take hours to complete and without fail it would lose steps, or a vwheel would fall off something else horrible would happen. It was just pure luck if I could get it do do what I wanted.

But I really wanted to be able to use the machine, so I started fixing the weak links. I got my belts tensioned correctly and places shrink tubing over them so they stopped slipping. I spent 3 hours adjusting the voltage levels to the motors, I upgraded the Z threaded rod to a ACME rod, I replaced the Nema 17 X motor with a Nema 23, I replaced all the eccentric nuts with eccentric washers and used nylock nut to tighten them down. I put blue locktite on every nut. I leaned how to square the z axis to the waste board, I used a flycutter to level my waste board.

Then I started on the software and bought Vcarve and learned how to use it. Just having the right software made as much difference as all the hardware mods.

So yes the machine is not ready to use after you tighten the last nut. And yes you will need to spend some more money to get it working the way you want it to. But that is what is really great about this hobby machine. You can upgrade it, You can improve upon the original design.

But it takes time and effort to get it working right. But you bought a $1,000 machine knowing that a turnkey machine would cost 6 or 7 thousand.dollars. So if you can stick with it and tune the machine correctly it will be capable of reliably doing what you want it to do. If you do not want to spend the time, then this may not be the machine for you.

Also you really should not leave the machine alone when it is running, especially before you have it tuned and working reliably. It is capable of starting a fire and with the issues you have described that is even more likely.

1 Like

Is there a link to a procedure for calibrating the voltage levels to the motors? I just upgraded to the X-Controller and need to calibrate everything again.

But to be constructive, is there no CNC controller interface/software that has any sort of tracking, auto calibration,self correcting, monitoring accuracy features?

Probably not until maybe the industrial class machines. I was just as surprised when I got a 3D printer just how dumb they are. They follow the directions they are given and every little move is predetermined when the gcode is generated. Once the CNC/printer is running it’s all dead reckoning. Most 3D printers can’t even tell if they’ve run out of filament.

Well I imagine the problem is not that big of potential problem on a 3D printer as there is rearly an issue of movement being obstructed, so it is possible to operate with predetermined intructions, while on a carve you are bound to run into a litle snag here and there, and then your doomed. But yeah, refinement is obviously still far from a reality unless you dosh out the cash in both lines of work, funny thing is i went with CNC machine after writing of the “cheap” 2k ish 3D printers for serious use, as “cnc has been around for a long time, so it must have gotten past the baby steps by now”

I thought the same thing, but have found the hobby level of open source hardware and software is a lot farther along in the 3D printer world than CNC.

Functionally, the machines are pretty much the same, but I have found the printer hardware to be much more reliable and the 3D printer software/workflows much more hashed out. I built my first printer in a week and was printing custom parts within two days. It’s taken many months to get the CNC built, and I expect a couple more learning the workflow well enough to create anything worth while.

Though, I will say that if you gave up on a $2500 3D printer, you are not going to be happy with a x-carve. Those $2500 printers would be more equivalent to a $7K - $10K CNC machine.

Yes it has. It costs big $$$$ for those machines. The hobby market has some catching up to do, but you can buy in for about $1000, not $100,000.