I’m brand new to the CNC world and am interested in a LR4 to build plywood furniture. At this point, I prefer the idea of ordering a turn-key machine from North Woods.
I’m looking for recommendations on reading materials or videos that will teach me how to use the controller from square one. I.e. once you have your file in Vcarve ready to be cut, what steps are required before you place the plywood on the table and press “go”.
I hope I’m not oversimplifying the process and appreciate your input. Thanks in advance!
Well, there are a few parts to this process. I don’t have much of an answer for reading materials or videos, but I think it helps if you can understand the big picture.
Building the Machine
This can be done yourself or by ordering one. You would be in good hands with @Jonathjon at Northwoods. The only downside of that is that you don’t get the experience of building the machine. At some point, you’ll have to fix something and that is easier if you were the one to put it together. That said, we’ll help you out. We’re a friendly bunch. I’ll also note that ordering one built isn’t completely “turn-key” since that gets you the gantry, but you still need to purchase rails and build a table.
Controllers/Firmware
The generally recommended controller here is the Jackpot which runs the FluidNC firmware which is largely GRBL compatible (that matters later). I believe all the Northwoods builds come with the Jackpot v3. FluidNC is generally used wirelessly and the controller gets mounted in the beam.
Software
There are a few categories of software here.
CAD / Design
This is where you design whatever you are building. This includes CAD software like Fusion, Onshape, FreeCAD, etc. It can also include other design software like Inkscape. You mentioned VCarve which is included here as well and other similar software like MillMage, Aspire, etc.
Assuming you are mostly doing 2D projects, these will output DXF or SVG files. 2.5D is also possible via STL files.
CAM
Once you have a DXF or SVG file, it needs to be converted into gcode that the controller can run. This is where the FluidNC/GRBL comes into play. The CAM software needs to generate that “flavor” of gcode. This includes software like EstlCAM (which is popular here). I believe VCarve also does CAM but I’m not sure. This also include software like Fusion, FreeCAD, MillMage and Aspire.
Each line/shape in the source file needs to be converted to some type of operation. You need to set depth of cut per pass and set feeds and speeds. This will depend on the machine, how fast the bit is spinning, what size and type of bit it is, and the material. You can read some more details on the milling basics page.
At the end of this stage, you have a gcode file that can be run by the controller.
WebUI/Sender
So, how do you get the gcode file and run it? With FluidNC, you would typically use the WebUI which is a webpage that you can use to control the machine. You can upload the file and run it from there. When you do it this way, the controller is directly running the job off it’s SD card. But, there are other things to be concerned about before you just run the file.
There are also a collection of GCode Senders such as Universal GCode Sender (UGS). Many connect to the controller via a USB cable but UGS can also do it wirelessly. These work differently because they send the gcode one line at a time to the controller. Some people like these but I generally avoid them.
Running a Job
This can vary but this is the general idea. You can also refer to this video from Ryan:
That video is using a different controller with a touch screen on the gantry. The only thing in this video I would be careful about is moving the gantry by hand like he does in the beginning. If you do this too fast, you can damage the controller board. I just jog the machine instead.
Positioning
Home the machine - The home action will position the machine in the bottom left corner and set the origin position. This is a repeatable process to the exact same location using the limit switches on the machine.
Position and Zero - This is setting the starting point for the job which may not be at 0,0. So, you jog the machine to the starting position and zero the machine (which makes that position the new 0,0).
Workholding
Obviously you need to place the material you want to cut. There are a few options for doing this. The easiest is just to directly screw it into the spoilboard but there are various other options with T-tracks or dogholes with some kind of clamp. Be aware that you don’t want to hit a clamp or a screw.
Start the job
Usually, you would install the first endmill prior to running the job.
As mentioned before, you can do this from the WebUI. It can also be done from a pendant which can be wired or wireless depending on which one (I am the developer of FluidTouch - Wireless FluidNC Pendant). You do not need a pendant and I would recommend at least waiting until you become more comfortable with your machine before getting one. There are enough things to learn.
Depending on how the gcode is generated, you will usually pause and have you probe the Z height. This means that you connect a touchplate and it moves down until it touches. From that we get an exact Z height. This isn’t super critical for through cuts because you give yourself some leeway, but anything where depth matters, this is important.
After that, you turn on the router and resume the job.
While the job is running, you must be monitoring it at all times. If it gets stuck in a spot or something happens, it can very quickly start a fire. If there is a tool change, you will need to stop the router, change the bit, re-probe, and resume. You also need to be prepared in case you need to stop it.
Also from a safety standpoint, be aware of the dangers of dust. This is no different than normal woodworking.
Only rail needed is the one Y rail. Everything else is included. I just cant ship a 10’ piece of EMT. And that is correct on the table as well. I do my best to make it as easy as possible though. And there are plenty here that can help you get a table set up real easy.
Correct. As of the last few we have transitioned from the JP1 to the JP3 now.
Jason, first of all, thank you so much for taking the time to give me a thorough response; greatly appreciated!
In response to some of your points; I didn’t mention I’m an intermediate-level hobbyist woodworker, so I have both a dust collector and shop vac/cyclone I could choose from for connecting to the machine. I’m excited about building the table, and was thinking of using dog holes, with plastic bench dogs on one of the X and Y edges, and Kreg in-line clamps on the another X and Y edges, also plastic - your thoughts on that? I have seen people using electrical conduit to build the rails - is that the preferred material to use?
Another general question, would a more complete system, such as the Yeti Smartbench or the Scienci Labs AltMill be significantly easier and less time consuming to learn how to use and gain proficiency with?
Thanks again for your time, I’m going to take a deeper dive online to some of the things you mentioned in your reply, mostly around the software and getting the gcode to the controller.
I dont think any other would be easier that the v1 system to learn on. GRBL is the underlying constant for all the cnc systems the others may be more ridgid. Meaning faster and deeper cuts per pass also magnitude more in cost. It the pockets are deep enough to justify reach. The lr4 is reaching into the big time players . You give up some speed ect but not decimal per dollar or monetary designation. If you need the ultra high precision you have to pay but for the dollar you can’t go wrong here.
The coolest thing about the LR is that you build the machine yourself and then know it in and out. It‘s also the best bang for buck, hands down.
I recently switched to a PrintNC after five years with V1 machines because I wanted more of everything. I would not have bought an Altmill, Stepcraft or build a Volksfräse because they are all, after all, aluminium extrusions. It is a step up, but not as significant for the price you are paying for those.
The LR is also more forgiving. You can crash it without breaking anything (expensive). Great for getting into the hobby and learning the ropes. If it‘s for you, you can still upgrade later.
1" EMT is what the LR4 is designed to use for its rails. The ones I build come with this already installed for the X beam and with the clips for you to installs for the Y rail. The large tubing really works well.
There are lots of clamping methods. I just screw directly to my spoilboard. The only part about your clamping thoughts that I’m not sure about is that you want the clamps to also have downward pressure. But at the same time, if they stick up at all, you need to be careful not to hit them with the router.
Well, we’re obviously a bit biased here. For the price, I think the Lowrider is fantastic. It’s also more forgiving to learn on. You can crash it and not break anything except maybe a bit. I’m not sure anything else is going to be easier. I think the software is the more difficult part and you have options there everywhere from free to thousands of dollars.
I can imagine with 3d printing where more expensive printers need less ‘tweaking’ to get good results right away that it would be appealing if the same was true of CNC machines.
I don’t think that’s the case but I have never used the more expensive CAD/CAM tools like Vectric so maybe they do make things easier.
Just no switching from cad to cam to visualizer its all together. I have been using vcarve pro for about 4 to 5 years I don’t remember. Before that I used inkscape and estcam and didn’t ever find a visualizer I liked. I still use universal gcode sender vetric does have a send direct to machine function.
Not sure if this question would be for you or Ryan… but could a machine with a 5ft X axis be built? Asking since I have only found 5x5 Baltic birch sheets, and I haven’t decided if I want to use Baltic birch or another type of plywood that’s made in 4x8 sheets
Before coming to your question: you can quote multiple people in one reply, then you only have one post and not a series of posts.
Most often you crash it by programming it wrong, for instance going to deep because of the wrong order of cuts or running into a clamp or screw or because you run into the workpiece when jogging or you run into something else on the table you put there and forgot, like a drill… So many options to mess up…
You can make it that wide. But it would not be advisable. The LR4 is pretty rigid at a 4’ cut width, but as far as I know no one has made one 5’ wide so we have no real knowledge of how it will run. My guess is a LOT slower. Something like that will need upgrades like Aluminum struts for sure I would think.
The lr4 is reaching into the big time players . You give up some speed ect but not decimal per dollar or monetary designation. If you need the ultra high precision you have to pay but for the dollar you can’t go wrong here.
