Entire List of X Carve Problems

Currently this X Carve has been an entirely wasted purchase. Complete waste.

I should clarify this isn’t the entire list, that’s much longer. But here’s the current things I’m trouble shooting

  1. Motors don’t lock tension under power. I can easily move the gantry on the X/Y axis

  2. Belts are loose but there is no video showing what proper tension should look like. The best I’ve found is tight, but not too tight. Awesome, thanks for that keen insight.

  3. No fine motor movement. Trying to move by .001 to zero the x and y axis and the machine just shakes the bit and doesn’t actually move, it just shakes the bit back and forth in both directions.

  4. Machined dimensions are wildly off from cam/cad dimensions. multiple mms

  5. The machine zero’s horribly and is based off of the center of the bit vs a jogged position normalizing for the cutters width.

  6. Easel’s zeroing system needs to be updated from zero machine to ‘set work zero’. The ‘confirm home’ needs to be labeled ‘confirm work zero’. The ‘use last home position’ needs to be changed to ‘use previous work zero’

  7. Z axis should be tossed in the trash. It wobbles under minimal torque as well as rotates along the X axis.

Whoever designed this machine needs to be fired if they haven’t been already. Impressively poorly done. Now I get to spend hours troubleshooting this crap and buying all new parts from other companies to make this thing even remotely operable.

1 Like

Regarding number 1. Turn off the x controller’s #4 dip switches off for each motor and change the $1=255.
Regarding number 2. I tighten the belts until they have some tension on them and raise up about an inch or so.
Regarding number 4. You need to calibrate the steps per mm for the machine to fine tune it. You probably can get accuracy into the +/- .005 if you work at it. It’s a math ratio of what you want to what you get for the $100, $101, and $102 commands. Read this:

As somebody who built a cnc and has had 2 commercially made ones, the X carve is a hobbiest machine. If you want more rigidity, speed, and accuracy you pay for it…in the neighborhood of $10K. I can say for the money that the X carve is a nice affordable entry level machine that is well designed and made. At that price point, compromises have to be made.


Thanks for the help! For a comparison in design quality go check out a Prusa 3d printer. The difference in cost, reliability, print/movement quality, detail, build quality, UI, UX, is really staggering. You’re right though, the x carve does seem like a diy hobby machine vs semi-pro though.

I just recently bought an ender 3 pro. It’s pretty much in the same ballpark and capable of much less than the X carve. A one trick pony so to speak. You have a smaller work area and shorter extrusions. This means more rigidity, you also have a nozzle as compared to the torque of a router… big difference…Really they are apples and oranges.


While yes, one uses a nozzle, and the other a heavier spinning cutter. And while one uses 12 inch extrusions vs 30 in extrusions. The differences I’m referring to are in quality of design when talking those things into account. And it’s not even a small level of wobble on the z axis, I can literally just wiggle the router left and right with minimal effort and it moves by mms. A shocking amount of play. The overall build is pretty trash from a systems engineering point of view and as a product designer, I can say that whole heartedly. This rig is off in the mms of movement while a $500 3D printer can move accurately to .1mm with no difficulty, code changes, etc. modern 3d printers have developed mesh leveling while inventibles is still trying the use the SAAS interface business model and hasn’t dialed in simple work zeroing. Every part on this thing turns out different by over a mm, sometimes 3-5. Just wildly poor quality

What I’m reading here is that you don’t have your machine properly assembled and for sure you don’t have it calibrated.
When you do get it assembled correctly, the Z axis will not wobble as you explained, and the overall accuracy of the machine will be within ~.005 - .010" (plenty close enough for wood working)
While I don’t know your background, I have been in the Machine tool, Tool & Die, Engineering
for 47 years, and I can assure you that I know quality.
That being said, yes, I have upgraded my machine also, new Z axis, cast aluminum bed, taller Y axis plates, even added a laser.
We must keep in mind that this is a hobby machine and not a commercial machine.
However, there are many people (including myself) that do in-fact use it to make money.
(mine paid for itself within 6 months)
Instead of spouting off, as a new member you should do some research here on the forum and ask questions.
Yes, it can be frustrating at times, but there are many people (including myself) on here that are more than willing help and guide you.
For the amount of money that you spent, this is an excellent machine and if you do have “legitimate” quality issues, you will also find that Inventables is an excellent company to work with.
Welcome to the forum.


There are thousands of quality projects in the Projects pages produced by hundreds of X-Carve users. All of them have navigated their way to success through perseverance, support, youtube videos and this forum

The X-carve is capable of producing quality work once properly adjusted. It is also popular to modify the X-carve just like the 3D printing community likes to modify their printers with special fan shrouds and spool holders etc.

Some of the issue is the small sales margins X Carves generate for Inventables. Another aspect is that Inventables basically competes against Shapeoko in this size/price range. The Sainsmart toy CNCs at $250 cannot be seriously compared to the X-Carve. The Nextwave Automation HD5 or the Axiom Iconic start at $5K. The Legacy Maverick is 3’ X 5’ and sells for $17K. Some of us have other CNCs, and these are not free of adjustment issues.

Inventables does not paint you into SaaS. You can totally use Easel to create G-Code that you then run from a G-Code sender and even run that code on other CNCs. Likewise, there are people who use software like Vectric and run their X-Carves entirely without Easel. There are popular companies who do lock you into SaaS such as GloForge lasers and Cricut vinyl cutters. Even Microsoft is trying to migrate Excel, Word, and PowerPoint to a cloud subscription model. Going the SaaS route allows companies to see how their products are actually being used so that they can make sure the popular features work correctly and are easy to use.


@RansomAllison - you might want to consider getting a Onefinity CNC then. They just launched a pre-sale and they seem to have a solid machine. Much, much easier to assemble – no belts, and more rigid. But, all that said, the advice given here is sound. It does not seem like you have the machine assembled properly and you do not have your expectations set properly. As someone who owned a Shapeoko, and owns an X-Carve, Prusa i3 MK2, and CR10, the build quality is the similar. The level of tinkering, well, that’s a different story. The stresses on a CNC far exceed a 3d printer; with a few years under my belt with all the above machines, you can’t simply jump into CNC and expect excellent results like 3D printing. It would be lovely, but isn’t the case. YMMV.


1 Like

For me, it’s the whole package at this price point that is such rubbish. It looks like the company stopped making tech upgrades in order to cash in on cloud sales focus. Since you own a prusa you can attest to how rapidly their hardware and software improves. With prusa there is a constant stream of tweaks a huge movements on the hardware, software, and UX/UI for a far better price. Now that I’ve spent 10+ hours going through and tuning this thing to get it in a usable state, which at this price point is just trash. The whole design behind it is just booty. It seems like the makerbot of cnc.

1 Like

Also, I just checked out the onefinity and that’s the exact setup I was going to recommend over this terrible v wheel design! I’ll probably get or just make one like that.

If the X-Carve doesn’t meet your expectations, take a look at the OpenBuilds machine kits. I’ve had two of then. They are very solid machines at their price points.

Bobscnc works fine for me. I use easel for being free. I still use V carve buts but tell it I am using 1/32 bit and adjust by 0.025 and get great carves and scripting.

User error


Sound to me like you rushed the assembly, didn’t do any calibration steps, didn’t read any of the hundreds of post explaining most of your issues on here, facebook or YouTube and now blame the machine. There are many happy users of the machine who did their research, assembled the machine, took the time to tune the system and produce parts with no issues.


I have 3x Prusa MK3 machines and an xcarve. I work in the film industry doing visual effects. I recently just finished Jumanji (if you like it cool, if you don’t, I’m sorry :frowning: ) I have spent the last 25 years making things both digitally and practically. I’m not anywhere near the amazing level of many of the people here in the forum but What I’ve read from them has brought me back from the brink many times and saved my relationship with my machine.

I understand your frustrations about the design. My background is design. I relate to your comments and agree about how Joseph at Prusa is always bringing out updates and fixes. But you are part of this design improvement now on the x carve. You can have a say here and suggest things and the forum will listen and I’ve seen the team at inventables respond and take onboard the best ideas.

Look up the post I did about hating the x carve solidly for 18 months. Like hate. Not your Girl guide bunch of fluffy flowers and dancing unicorns level of forum frustration but kicking kids off a cliff into a burning pit of lava and hornets level hate. Like lacing door knobs with Covid at an orphanage hate. So what I’m saying is I get you. I was you. I was an angrier version of you with all the same “I’m a designer, head me roar” pomp and bluster. Turned out a grub screw had fallen out of my x axis and basically shagged my whole machine. Up until that point is tried everything. Belt tightening, stiffening, everything. When I fixed the grub screw issue my otherwise massively calibrated machine was suddenly golden. (Thank you forum gods for all the amazing posts and advice).

So I agree that the design could be better. Tell me something that couldn’t be better. Don’t list the Prusa as the best thing ever either. I owned the MMU which was a tank of rotting ■■■■ and blood filled vomit for a long time. Took Prusa a long time come up with something “slightly” better. So they are far from perfect. I run my three printers almost 24/7 now and have calibrated them to 1/4 proton so life is great. But that took a while.

The X carve is a hobby machine and didn’t cost the earth. Yeah $2000 isn’t nothing. But if you are poor, CNC machines are lower on the list than most other essential items like food, clothing and sex toys.

So go through the process like the rest of us, know that we all went through some form of this as well and that we have all ended up for the most part with a machine that can cut beautiful pieces.

Shit really gets cool when you start combining the Prusa and Xcarve together and making combo projects. That’s for a different subject. But welcome to the forum. Look up my post on nearly smashing the xcarve to bits with a sledge hammer.



Everything you listed is improper set up alignment.


amen , some people expect a Cadillac when they only paid for ford pinto ,

1 Like

Another guy with a couple Prusa printers here. The X-Carve is what it is, if you want a beefier production system then you’re looking in the wrong place. As others have noted, the majority of your problems are both atypical and directly traceable to issues with assembly. That’s OK, because that means you can fix those issues if you chose to address them in a constructive manner.

If you don’t think Easel is the greatest software ever, then you’d be in good company and also in luck, because there are a crapload of GRBL-compatible solutions available for you to choose from.

I’d suggest you keep plugging away at this. Take one of your problems, ask around the forums here, get it nailed down, and proceed to the next. Just like a 3D printer kit or anything else sufficiently complicated, it can be frustrating when you’re faced with several problems at once.

A lot of people here have made this work out for them, and better still, they’re here to help you do the same. That’s another Prusa-like advantage to the X-Carve: a solid community that will help you get your machine working like you would expect (within design limits).


Since your last post was 6 days ago you probably won’t read this but someone in the community will.
I purchased my X-Carve about 3 weeks ago
After I built it, I made the little plaque that is in the Easel demo
I change the font to a script, and this created some very small center sections in the E’s, Y, and the little triangle at the top of the Easel Logo - They broke off ( I thought it was because of the quality of the birch plywood that came with my Kit )
I decided to check the alignment of the machine, and double-check that every axis is perpendicular to another
Before now I didn’t have a 3D printer or CNC machine, I build things, so I watched this video series https://youtu.be/xJh1pMIxqJ4 before the machine even arrived
I’m glad I did, because I only had to make one adjustment to fix the problem I was having, and it was the front and back of the Z-axis
Basically the router bit was going in at an angle, more towards the back ( / not | )
I fixed this by twisting the top of X-access rail towards the back of the machine
I discovered this by accident because I ran it without installing the dust collector
When I installed the dust collector I noticed the front of the dust collector is lower than the back
I immediately thought; oh crap the router is crooked it’s leaning forward.<( I wanted to put an exclamation point here, but I don’t want to yell :slight_smile: )

Moral of the story:
Completely build your machine before you use it, and double-check that every axis is perpendicular to another

-Ken ( Steps off his soapbox, quietly)

PS: The dust collector leans forward by Design. The locking screw, at the top of the mount, pushes it away from the X-axis. There is a bit of play in the rails and dados of the mount. This is not a problem because it makes it easier to adjust


Very well said!