Many of us have spent most of the year on the COVID-19 lockdown, what must we show for this? Has the Netflix queue made some progress? Maybe it's a (slightly) cleaned garage or workshop? If you are [Bob] of Make Stuff, you won't: he has been working on a fully customized CNC router for the past nine months, which is large enough to hold a whole piece of plywood.

The build was recorded in a series of nearly ten YouTube videos, the first of which was released as early as January 2020. Seeing [Bob] go to the steel plant to get the frame components, but can't see the mask to remind him how long he has been working on this project. If anyone wants to follow in his footsteps, he also compiled a comprehensive list of materials on his website. The price is only slightly less than $4,000, which is certainly not a budget build. But when we talk about machines of this size, nothing is cheap.

Even if you do not build your own version of the router, it is impossible to look at the build log without understanding the possibility of such a machine. In the previous video, we even saw some self-replicating actions, because the giant CNC cuts parts for its own electronic device casing.

You can see from the video that [Bob] is proud of his creation (as a matter of course) and shows the audience every victory along the way without shyness. Even if things don't go according to plan, he can learn from them when he explains the problems and how to solve them in the end.

Of course, we know that a homemade CNC router does not need to cost thousands of dollars or take up as much space as a pool table. Ordinary Hackaday readers may not need such a monster, and even if they do, there is nowhere to keep it. But that doesn't mean that we can't jealously watch what kind of projects [Bob] uses such incredible tools in his arsenal to make.

4k is a good deal for a router of this size. It’s even too cheap and the frame doesn’t look so sturdy

Those legs definitely need diagonal bracing. That little dog writhing around. But it is easy to fix.

Fix the plywood to 3 sides, can you still use all the space below? And take it apart when needed.

In my life, I never thought that anything made of wooden parts would be "heavy".

Straight legs like that are crazy. Can not. Fix it by welding cross braces between the legs and the bottom of the table. Otherwise it will bend like a fish.

I swear-I really don't want to be mean, it's just a simple design basis. You need rigidity and quality to use large gantry routers for processing. Wood will warp with humidity and temperature, and it is not suitable for parts on large machines that require high precision.

You are well covered. He should cut those wood parts with aluminum. Steel is better. Obviously, this is a machine that can be easily repaired. It is actually good, but it may still be "heavy" now.

It's just a stepper driver, it's not a real structure

However, plywood has fewer problems than bulk wood. You can also boil wood in oil or wax to increase resistance to changes in humidity. Although completely agree with the legs, the machine may have a huge resonance peak, which coincides with a reasonable running speed.

He later added croosbraces.

I used about 500 dollars to build a similar CNC that is about 1/3 the size. Suitable for wood and 6mm acrylic board.

It may be entangled with metal, but this is not what I built it for.

Look at Maslow. It is a vertical CNC controlled router kit. A whole piece of plywood (or anything else) will be cut. Take up less space. The cost is approximately $1,000 (give or accept).

Are you using one? Maslow's kinematics is very attractive, but it also has some fatal flaws-such as how you can withstand a lot of pressure when you "dive".

Yes, I built one. They work, they work very well. They don't have any "fatal flaws", otherwise no one would build one.

The "sleigh" has enough weight. It is the plywood or MDF disc that supports the router. This is accompanied by two weights, and the angle of the cutting board (about 15°) is sufficient to apply sufficient vertical force to cut.

I like my Maslow, but I spend as much time at work as I use it.

You are wrong to do this. I built mine, calibrated it, and now I just use it.

Maybe your luck is better than mine, but my experience is that Maslow is a very sensitive device. I like to fiddle with it, but it can definitely tell you when you are in a hurry.

Yes. The only thing I can agree with is that Maslow put "slow" in the CNC wiring. However, it is cheap and powerful.

Like him, I can see room for improvement, such as using machines to cut electronic boxes. Once it is built, you can improve it and use the parts it can make. Just like every project, one step at a time, but it's done, when it does what he wants to do. If it meets the requirements of the builder, there is no need to overextend it. Before I build something like this, it’s great to be able to observe other people, preferably many others, to get new ideas, inspiration, motivation, and to help avoid learning things the hard way. It's not so much about copying, it's about improving the available tools and being creative. It is not so much what we have, as what we do with it.

He spent so much money on that router that he couldn't even afford a haircut.

This guy is doing cargo worship precision for his project. In any case, the steel beams supported at the ends will sag in the middle, and will sag depending on the way the ends are supported, so check that the quantity in one case does not apply to the other. The rope is the same: you can't pull it tight enough that it won't sag. Similarly, align the frame with the "playing cards" on the floor before welding, it will be different after welding, because welding will break it, and when the frame is moved and different loads are applied to it, the stress will change. Well-tuned airplanes are everywhere.

So he did all this in vain. If he really wants it to be straight everywhere, he must add some method to prestress the structure to resist the load and push it straight after assembly. Otherwise, all his measurements will be changed four times during the construction process.

A simplified example of a beam suspended between two points that can be pivoted freely:

y = (5*m*g*L^3) / (384*E*I)

m and I can be found in the beam catalog of the steel supplier. The E for steel is 210 GPa. Take a typical 40x40x4mm tube as an example: 4.2 kg/m, I in the basic unit = 1.25952e-7 m^4.

For an 8-foot beam (2438 mm), the result is 7e-4 meters or 0.7 mm (27 mm), which is the degree to which the beam sags only due to its own weight. Trying to measure its "straightness" with a feeler gauge, and it is bending over anyway, this is a meaningless exercise.

How to solve the straightness problem: Weld the two ends of the beam together, and drill two holes in the center of the bottom. Pass the long threaded rod all the way through the hole and start tightening. This will prestress the beam to resist gravity and bend it back. Of course, any load you put on the beam will cause it to sag again, because you just canceled the weight of the beam, not any additional load. When the gantry rolls on the beam, it will still bend downward.

More importantly: Assuming that the rope is tight enough to be "straight", and the person finds that he has a gap of 4 thou (0.1 mm), what can we infer? We now know that the beam should be bent by 0.7 mm, but this is not the case, so it must have been bent by 0.6 mm, and he happened to bend it like a bridge arch. Liang is not straight, he is just lucky. If he flips it, it will bend down by 0.7 0.6 mm = 1.3 mm.

This is why you don't believe your material. Either design the mechanical device so that it remains level after assembly, or design the machine so that it must be level anyway. For example, laser light is irradiated into the photodetector to follow the beam and continuously correct the Z-axis offset.

https://www.engineeringtoolbox.com/cable-loads-d_1816.html

In order for the rope to sag no more than 0.01 mm in the middle, we call it "straight" because it is smaller than the range you can reasonably measure. Assuming it weighs 1 gram per meter, you must hang 7.2 kg to pull it from the bottom. If it is a thicker rope weighing 2 grams per meter, you need 14.4 kilograms and so on.

It is impossible to pull the rope absolutely straight, because when the rope is a perfect thread, the leverage you get is unlimited. Offset and sagging require zero force, so you can only approach straight lines. So the only question is, will your rope pull?

0.01 mm is unrealistic for a gantry of this size, let alone necessary for wood. To put it in context, the flatness tolerance of the laboratory's AA grade 8'x4' platform is 0.0127 mm. Even if you spend more than $11,000 to buy a 3000 Kg, 16-inch thick pink granite slab, and install it in your heated garage at any cost, you still can’t make your gantry so flat.

Carpenters rarely use tolerances below 5, and joiners need their strictest tolerances, especially for visible surface connections.

Even though his rack is very flat, his floor is not. Increasing the thickness of the beam, reducing the length of the beam (possibly more legs), supporting stiffness and including adjustable feet are practical measures for improvement. Keeping it flat during welding (or grinding it down) is also a challenge that may need to be solved to achieve the required tolerances.

Yes, this is something you can no longer reasonably measure, so in this case it is as good as "direct". After all, your standard must be better than what you can measure with it.

That is, he did the wise thing and put the guide rail on the adjustable side, so it doesn't matter that the top of the final frame is not completely straight. This just means that the initial fine-tuning and adjustment is futile.

The difficulty is that the beam will also bend to the side, so now the problem is just transferred to another shaft, hoping that the linear bearing will not get stuck. It happens that both sides are bent in the same way, so what if the gantry has a banana-shaped track? Well, again, you have to tighten the frame in some way to remove the curvature, but first you have to find a way to measure it.

Otherwise, if you grind out some large objects on it, it will also bend.

Please be kind and respectful to help make the comment section great. (Comment Policy)

This website uses Akismet to reduce spam. Learn how to handle your comment data.

By using our website and services, you explicitly agree to the placement of our performance, functionality and advertising cookies. learn more