I am starting this thread to document my experiences with learning to use home CNC router for wood and light metal carving work. I want to share the resources that I will discover and what pitfalls and challenges I encounter with this technology.
The machine I'm working with is a Millright Mega V kit that costs a little over $2000 in the smaller size that I ordered. I chose this model because of the open sides and fronts that allow larger materials than the machine is capable of cutting. So basically even though this machine can only work on a 19"x19" area I can still fit a surfboard through it and carve up a 19"x19" design into the board. There are a lot of smaller options for a home CNC as well but this one struck me as being as large as I could reasonably accommodate and afford. While I would like to eventually mill out some aluminum molds and other metal projects the majority of my planned use will be with wood. This machine is supposedly rigid enough to mill aluminum and occasionally harder materials.
Assembling the kit in the garage. This kit was an easy assembly but the instructions were a little spartan and could use some refinement. If you have built a 3d printer or assembled a computer this kit shouldn't have any significantly new challenges.
I assembled this kit and did basic motion testing nearly a year ago but I only just got around to carving. My first point of procrastination was coming up with a permanent stand and home for the CNC router. I was very lucky my work had this strange, extremely tall, kitchen cabinet with a bullnose granite top that is just barely larger than CNC. We just wanted it gone at work and have been trying to rehome it for 3 years so this CNC project was a perfect match for this old abandoned cabinet.
In my eyes this is the perfect stand for a CNC, the granite top is nice and flat and very hard to ding or dent. Also the cabinet can be bolted to the wall providing increased stability. The cabinet top is about 48" off the ground so it's very tall compared to normal cabinet heights which are usually around 32" I believe. I cut cord passthroughs on the back left and right top corners and placed an old desktop computer workstation inside. I eventually set up the workstation and the CNC controller on a backup UPS Power supply to help provide surge protection and a soft shutdown in the even of a power outage. I have yet to program in an soft-shutdown features though so this is just theoretical. The router spindle will not be powered by the UPS so I am hoping to just use the UPS power outage signal to tell the CNC to pause the job and raise the head to get the spindle out of the work piece while it spins down from lack of power.
I removed one the top drawers to make room for the CNC's computer housing. This just houses an arduino with a stepper driver breakout but it does have these fancy aviation connectors that help it feel more substantial. I ended up using some scrap metal and wood to block the CNC computer in so i can slam on the emergency stop button and not send the entire computer deeper into the cabinet.
Here is a photo I took of the backside of the CNC computer so I could install the cables into the right slots without spinning the whole unit around. As you can see from this photo there is capability for a laser module add-on that I am not utilizing.
Here is current setup with a LCD monitor on a pegboard shelf above the unit. This allows me to use Universal GCode Sender, an open source software package, to manually control the CNC or send Gcode files with preprogrammed instruction.
When I first started with this machine it I kept having issues with it not being to "Home" itself. 3d printer nerds will be familiar with this as it's where the machine learns where it's tool head is positioned in X,Y, Z coordinates. The machines know how to move the motors any time, but it only knows where the machine tool head actually is once it has been "homed" which is done by sending the machine to furthest extents (or minimums) until it runs into mechanical switches that are pressed by it running into them. This tells the machine to stop as it's hit the switches and then the machine knows where it is and therefore where it can go before it hits the ends of the gantry and rails. My issue though is that the machine was not respecting the limit switches and would instead crash into them and attempt to keep going. This turned me off this project for months. I had plenty of other projects to do in the mean time but a large machine with incredible strength, that doesn't listen to it's owner or safety switches, is terrifying.
Eventually I ran out of projects though and it was time for the CNC to "get busy carving or get busy dying" as they say in Shaw-quarter-inch-Shank Redemption. Armed with a bit more familiarity with Universal GCode Sender I ran the machine setup wizard and found the Z-limit switch was not functional. This limit switch is the first to be tested during the homing process so it makes sense that it's failure would impede the entire process. Limit switches are just a mechanical way of making two wires touch so I removed the push connectors on the back of the limit switch and touched them together. Still nothing changes on UGS which tells me that the wire itself or the connection to the CNC computer must be the issue, not the actual limit switch. I chased the Z-limit wire back to the CNC computer and started disassembling the aviation connector. It did not take long to see the issue was a failed solder joint.
Failed solder connection on Z-limit was the culprint for my homing issues. I used a hot-air reflow gun to reflow the solder to the wire as I didn't have my portable soldering iron with me. Fortunately the aviation connectors can handle a lot of heat and didn't melt (much).
After that was fixed I loaded in a very small ball nose mill and some scrap wood to try my first test carve. This was a design I had made in Vectric Aspire as an absurd take on a "Live, Love, Laugh" sign. The phrase is popularized by the navy seals but the juxtaposition of the scenarios this phrase invokes in the context of a live love laugh sign make me giggle inside.
Survive, Evade, Resist, Escape
With new confidence and some arrogance I moved straight onto a real project, a sign for my mom's upcycled product line called "Sofrina Green Collections". The logo was designed by a professional a long time ago so I was able to import the high quality art file give it a try.
It was working at a great speed and depth, but it was taking some really aggressive cuts in some areas that seemed to make it stutter as it moved which seemed like a bad sign.
Eventually I lost the X-axis position and the S of the logo was repeated ruining the piece.
I cut a new piece and had it program a new tool path for the Gcode. I switched bits to a 60 Degree V-bit so I could utilize the prismatic effect it creates.
Hmmm, still not right.
Well at least the issues are the consolidated to the X-axis. I did some googling of the X-axis issues with MillrightCNC machines and it said first thing was to check that the X-axis carriage is secure. What an idea, ha! I check it out and notice that one of the four wheel is a little loose. These V-wheels are the same concept as what is used on an Ender 3 and many other 3d printers so this was not my first loose gantry wheel. In the case of V-wheels the ability to tighten them down to the gantry and ensure good surface contact and friction is by adjusting an eccentrically cut nut that is on the axis of two of the four wheels. By rotating the eccentric nut you are adjusting the axis of the wheel's distance from the gantry and can thus tighten or loosen the wheel's grip. I tightened mine down and was able to carve again with the same file on the other side of the wood. I will laminate a thin layer on the back (to cover the failed carving) and we can all pretend I got it right the first time.
A successful V-bit carve of my mom's logo.
The machine I'm working with is a Millright Mega V kit that costs a little over $2000 in the smaller size that I ordered. I chose this model because of the open sides and fronts that allow larger materials than the machine is capable of cutting. So basically even though this machine can only work on a 19"x19" area I can still fit a surfboard through it and carve up a 19"x19" design into the board. There are a lot of smaller options for a home CNC as well but this one struck me as being as large as I could reasonably accommodate and afford. While I would like to eventually mill out some aluminum molds and other metal projects the majority of my planned use will be with wood. This machine is supposedly rigid enough to mill aluminum and occasionally harder materials.
Assembling the kit in the garage. This kit was an easy assembly but the instructions were a little spartan and could use some refinement. If you have built a 3d printer or assembled a computer this kit shouldn't have any significantly new challenges.
I assembled this kit and did basic motion testing nearly a year ago but I only just got around to carving. My first point of procrastination was coming up with a permanent stand and home for the CNC router. I was very lucky my work had this strange, extremely tall, kitchen cabinet with a bullnose granite top that is just barely larger than CNC. We just wanted it gone at work and have been trying to rehome it for 3 years so this CNC project was a perfect match for this old abandoned cabinet.
In my eyes this is the perfect stand for a CNC, the granite top is nice and flat and very hard to ding or dent. Also the cabinet can be bolted to the wall providing increased stability. The cabinet top is about 48" off the ground so it's very tall compared to normal cabinet heights which are usually around 32" I believe. I cut cord passthroughs on the back left and right top corners and placed an old desktop computer workstation inside. I eventually set up the workstation and the CNC controller on a backup UPS Power supply to help provide surge protection and a soft shutdown in the even of a power outage. I have yet to program in an soft-shutdown features though so this is just theoretical. The router spindle will not be powered by the UPS so I am hoping to just use the UPS power outage signal to tell the CNC to pause the job and raise the head to get the spindle out of the work piece while it spins down from lack of power.
I removed one the top drawers to make room for the CNC's computer housing. This just houses an arduino with a stepper driver breakout but it does have these fancy aviation connectors that help it feel more substantial. I ended up using some scrap metal and wood to block the CNC computer in so i can slam on the emergency stop button and not send the entire computer deeper into the cabinet.
Here is a photo I took of the backside of the CNC computer so I could install the cables into the right slots without spinning the whole unit around. As you can see from this photo there is capability for a laser module add-on that I am not utilizing.
Here is current setup with a LCD monitor on a pegboard shelf above the unit. This allows me to use Universal GCode Sender, an open source software package, to manually control the CNC or send Gcode files with preprogrammed instruction.
When I first started with this machine it I kept having issues with it not being to "Home" itself. 3d printer nerds will be familiar with this as it's where the machine learns where it's tool head is positioned in X,Y, Z coordinates. The machines know how to move the motors any time, but it only knows where the machine tool head actually is once it has been "homed" which is done by sending the machine to furthest extents (or minimums) until it runs into mechanical switches that are pressed by it running into them. This tells the machine to stop as it's hit the switches and then the machine knows where it is and therefore where it can go before it hits the ends of the gantry and rails. My issue though is that the machine was not respecting the limit switches and would instead crash into them and attempt to keep going. This turned me off this project for months. I had plenty of other projects to do in the mean time but a large machine with incredible strength, that doesn't listen to it's owner or safety switches, is terrifying.
Eventually I ran out of projects though and it was time for the CNC to "get busy carving or get busy dying" as they say in Shaw-quarter-inch-Shank Redemption. Armed with a bit more familiarity with Universal GCode Sender I ran the machine setup wizard and found the Z-limit switch was not functional. This limit switch is the first to be tested during the homing process so it makes sense that it's failure would impede the entire process. Limit switches are just a mechanical way of making two wires touch so I removed the push connectors on the back of the limit switch and touched them together. Still nothing changes on UGS which tells me that the wire itself or the connection to the CNC computer must be the issue, not the actual limit switch. I chased the Z-limit wire back to the CNC computer and started disassembling the aviation connector. It did not take long to see the issue was a failed solder joint.
Failed solder connection on Z-limit was the culprint for my homing issues. I used a hot-air reflow gun to reflow the solder to the wire as I didn't have my portable soldering iron with me. Fortunately the aviation connectors can handle a lot of heat and didn't melt (much).
After that was fixed I loaded in a very small ball nose mill and some scrap wood to try my first test carve. This was a design I had made in Vectric Aspire as an absurd take on a "Live, Love, Laugh" sign. The phrase is popularized by the navy seals but the juxtaposition of the scenarios this phrase invokes in the context of a live love laugh sign make me giggle inside.
Survive, Evade, Resist, Escape
With new confidence and some arrogance I moved straight onto a real project, a sign for my mom's upcycled product line called "Sofrina Green Collections". The logo was designed by a professional a long time ago so I was able to import the high quality art file give it a try.
It was working at a great speed and depth, but it was taking some really aggressive cuts in some areas that seemed to make it stutter as it moved which seemed like a bad sign.
Eventually I lost the X-axis position and the S of the logo was repeated ruining the piece.
I cut a new piece and had it program a new tool path for the Gcode. I switched bits to a 60 Degree V-bit so I could utilize the prismatic effect it creates.
Hmmm, still not right.
Well at least the issues are the consolidated to the X-axis. I did some googling of the X-axis issues with MillrightCNC machines and it said first thing was to check that the X-axis carriage is secure. What an idea, ha! I check it out and notice that one of the four wheel is a little loose. These V-wheels are the same concept as what is used on an Ender 3 and many other 3d printers so this was not my first loose gantry wheel. In the case of V-wheels the ability to tighten them down to the gantry and ensure good surface contact and friction is by adjusting an eccentrically cut nut that is on the axis of two of the four wheels. By rotating the eccentric nut you are adjusting the axis of the wheel's distance from the gantry and can thus tighten or loosen the wheel's grip. I tightened mine down and was able to carve again with the same file on the other side of the wood. I will laminate a thin layer on the back (to cover the failed carving) and we can all pretend I got it right the first time.
A successful V-bit carve of my mom's logo.
Last edited:
