Jonathon Duerig published a new build:

Read more about this build...

 

@GrayUK Thanks. I just updated with todays progress. The table/shelf base structure is complete. I'm at last starting to put together the actual pieces of the router.

-D

 

More progress today. A first look at the X-axis and the electronics base board. No custom plates for the X-Axis gantries. Instead, I use a vertical piece of extrusion which will attach to the standard double wide C-Beam gantry plates at the bottom and the C-Beam of the X-Axis at the top.

-D

 

The mechanical and wiring parts are now complete. Now just to test things out. Also it is time to start thinking about dust collection and workholding. I have some ideas to try out there.

-D

 

I managed to make my first cut today. The hybrid pressure foot and dust shoe seems to work pretty well. The vacuum doesn't collect all of the debris, but it does seem to do a great job at keeping it contained. The main problem is that I need to increase the clamping force of the pressure foot to make sure that even small pieces stay where they are as I cut them out. One of my main goals of this is to avoid the annoying and time consuming task of breaking the tabs off of pieces.

-D

 

And this build is complete. I'm actually producing parts. The main pieces seem pretty standard. But here are the things I think work best and the lessons I feel I've learned:

- For a large cnc router, build it into a sturdy table. The best place for the electronics is underneath the spoilboard since that will be the place that gets the least swarf/dust even if it isn't enclosed.
- Build with all the 'upgrades' you plan in mind. Homing switches are essential for production work. Drag chain is not optional at this size. Dust collection is essential, but it is a win if it contains the mess for later cleanup even if not all the dust is vacuumed up.
- The router is only as useful as your workholding system. The difference between an ad-hoc workholding system and a well thought out one is the difference between a hobby machine and one that can be used professionally. Speed and other 'professional' features don't matter as much because that is cheap table time while workholding and cutting out tabs is precious human time.
- For cutting small parts out of thin (1/8") material, a pressure foot is worth its weight in gold. On larger parts, other clamping mechanisms work better. But avoiding tabs means avoiding a lot of human effort.

-D

 

@LukasK I can take a few more photos of the connection. I went a bit overboard on it. I had a double L-Bracket on either side, an L-plate where the two are flush, and a single L-Bracket screwed to a three-hole strip plate opposite. There has been no flex on it as far as I can see. The flexing of the machine is entirely in the wheels in the X-axis gantry.

The clearance height between Y and X profiles is driven entirely by the presences of the L-Bracket. So it needs to be at least 40mm. You could have as little as 20mm clearance if you were willing to turn the L-Bracket into a strip plate. Since the connection between them is so secure, that might be reasonable.

The biggest point of flex in this machine is the wheels on the X gantry plate. So if I had it to do over again (and I might switch to this in the future), I would try using a pair of X-Large gantry plates instead of the double-wide plate there and end up with something like this:

http://cdn1.bigcommerce.com/server2...uploaded_images/untitled.191.jpg?t=1461787092

The doubled up wheels running along the outside of the C-Beam with larger distance between them would reduce the amount of flex significantly. And it would let me turn the C-Beam around so that the interior section with the lead screw would be less likely to get dust or grit from the cutter.

I look forward to seeing your build. And feel free to ask me about anything else that I've done or my experience with it.

-D

 

@LukasK I don't see a lot of advantages in X-Large plates on the Y-axis. If I wanted to have a very broad base of support, I would use two normal gantries or two double-wide gantries instead. For cutting wood or plastic, that seems like overkill anyhow.

Since the X-Large plates have wheels that run on the outside, they would interfere with support crossbeams connected underneath. I think having several support crossbeams is essential to making a solid base.

And since the Y-Axis supports the X-Axis on both sides, there is not nearly as much twist, so that benefit doesn't seem as great.

-D

 

I used this one:

CNC Pressure Foot Clamping Attachment - Clamp and Cut Parts on Your CNC Machine Without Using Tabs or Double-Sided Tape - WidgetWorks Unlimited

However, for my rev 2, I replaced their aluminum mounting plate with a 3d-printed vacuum attachment. The only parts I'm using from their kit right now is the actual foot, the springs, and the two screw/axles that the springs sit on.

I'm currently doing a major update of my machine which includes having two independent spindles. I was planning on 3d-printing a foot for the second spindle. What does your 3d-printed foot look like? I might be able to suggest a fix that would be a lot cheaper than the pressure foot I bought.

Also, I think that maybe one of the nylon or alloy 910 materials might be a better material for the foot part than ABS or PLA.

-D

 

I've attached my pressure foot files if you want to look at them. These two parts come together to form the part that mounts to your Z-axis/spindle. There is a vacuum mount, a hole for the spindle pieces to go through, and it is screwed together with four M5 screws. The wing on either side is for mounting it (I used L-Plates and attached it to T-Slots next to my spindle mount). The holes on either side are for the screws down to the foot itself. It uses two posts (rather than four posts like you have). The big complaint I have is that between only having two posts and the wings being so thin, it tilts too easily. Not much is holding it vertical. As I said, the actual foot is still the one from the kit that I got.

Looking at your design, there are two big things I'd fix. First, you should use three posts instead of four to connect the foot to the top. Three points define a plane. So I think that your springs and such are more likely to jam with four posts. When they jam, they exert a lot more pressure than just the spring and that will cause you to lose steps. Stronger springs also make it more likely to jam, and the closer the tolerances of your posts to their casings, the more likely it is to jam. My design is actually a bit too loose as I said, but the chances of jamming are zero.

The other thing is that your foot geometry is opposite of how it should be on the bottom. You want a fillet or chamfer at the bottom so that if it hits a pieces of debris or anything which is slightly proud, that collision is at an angle and so it won't catch. Instead, your foot geometry has a thin 'skirt' that seems certain to catch on lots of stuff. So make sure to get rid of the 'skirt' and instead make the bottom edges less severe. Since the first layer when 3d printing must be super-thick, I'd sand or file around that bottom edge after 3d printing to make sure that it is a gentle curve or angle.

The goal of a pressure foot is to minimize friction between the foot and the top of the material while maximizing friction between the bottom of the material and the table. A part of me wonders if there would be a way to add an extremely non-slip material as the top of the table. While having the pressure foot itself consist of three little roller bearings: 5/8 in. Roller Ball Bearing

-D

 

I may have misinterpreted your pictures. If your skirt is designed so that things can slip under it, then you are most likely experiencing jams which are causing more forces on your machine than you intend.

The advantage of mounting some place that does not move up/down is that you can have a constant clamping force. The disadvantage is that it won't automatically lift clear of the material at the end making mounting/unmounting more fiddly.

I should note that when using a pressure foot on wood, it is more effective to use a down-spiral bit that forces the sawdust down and packs it. This makes the piece more stable because it is partially supported by the packed sawdust on the last pass. And a vacuum that sucks away sawdust works counter to this. I'm working with plastic and if I tried to do it that way, it would melt. This is one reason why I use a vacuum attachment and upcut bit. But I probably get more defects than an equivalent setup with downcut no-vacuum setup would on a sheet of wood.

I'll try to post a photo here of my foot in operation when I get a chance.

-D

 

As part of my rebuild of the CNC router, I'm redesigning my pressure foot system. This time, I'm not keeping any parts or pieces from the one that I bought. I've 3d printed the new foot itself. When I get the parts for the other pieces, I'll try the whole thing out.

I've attached the STL for the foot. My idea is that I can use three half-inch shoulder bolts (with 3/8" threads) as posts. The hexagonal cutouts on the bottom are just big enough for 3/8" nylock nuts. Using normal nuts or jam nuts, it might be possible to make the foot thinner.

I also had an idea for a new way to capture nuts that I might try at some point that uses rectangular T-Nuts like these: Pre-Assembly Short Nuts -For HFS8 Series Aluminum Extrusions-｜MISUMI｜MISUMI USA

-D