I've been struggling to get precise cuts on my workbee. Cutting symmetrical things is proving to be more of a challenge than I thought (Example below) You can see there are some inconsistencies between the two pieces, even though they are very similar in form. I've calibrated all axis through Openbuild Control a couple of times to make sure that wasn't the problem, I'm still stumped on how to fix this issue. Any help on how to fix this would be greatly appreciated!
Went to check if the machine was out of square and it was by1/8 of an inch, took me a while to get it all square. After that I ran the same test I used in my first post, unfortunately I'm experiencing similar results. I'm not too sure what to try next
Ok, well at least you fixed a problem Next thing to check is backlash. We will also need to knows all the details of how you are cutting these out. Bit used, speeds/feeds, depth of cut, alignment of pieces on your machine bed, etc....
I had a problem with backlash before which was the result of the pulleys on the stepper motors being loose. I used loctite on all the grub screws so that should have fixed it. Maybe there's still backlash? I'm using a 1/4" upcut carbide endmill to cut these out For the speeds and feeds, I'm referring to Alex Chambers Feeds and Speeds chart he posted on the forum. which dictates 100" Feed and 50" Plunge. However, I'm not sure if those numbers are for the leadscrew driven machine (I have the belt-driven version). That could be the problem but not too sure. I set the depth of cut to 1/8", I had it set to 1/4" before but I was worried I would damage the machine by pushing it too hard. I align the work with a lip I cut into my spoil board, I don't really have an elegant solution for alignment
For alignment, I meant what way are the pieces facing when you cut them so we know what cuts were made in what direction. There's a lot that goes into this stuff. I have a hybrid of a workbee that has screws on the Y and a belt on the X. Your numbers seem reasonable. Are you sure your belts are tight enough? Play in the wheels can cause issues as well. Undersize bits can cause mating of parts problems. Lots and lots of things. More pictures especially ones since you've squared your machine (I keep looking at your original picture and all I see is a machine that's out of square. lol). Mark the X and Y directions on the parts. More test cuts with lower feed rates maybe. Test cuts with the parts rotated in a different direction perhaps. I've had my machine for almost 2 years and I'm always chasing something down. Learning how to quickly ID where the issue may lie is a valuable asset to have and one we can teach you, but we need as much info as you can provide.
Jose, did you make a test file cut on your machine? You know, something with two different size circles and squares, all on the same center, with maybe two diagonals? It will tell you if you have a problem and if so, on which axis. Mark on the board with a pencil where the X axis and the Y axis are. BTW, no need to go deep. Just a shallow pass, maybe 1.5mm - 2mm deep. The dimensions are just a suggestion and can be anything you like.
Decided to make new test cuts with the recommendations you all gave me. Everything was measured in inches with calipers These are the measurements the shapes were meant to be Outer Circle 6.5" Inner Circle 3" The Inner Square 1.5" Outer Square 4" All the cuts were set as outer cuts. From what I measured it's clear that the Y-axis is off, making cuts shorter than they should be. But the Outer circle is bigger on the X-axis by 1/32 of an inch, with the Y-axis being off by a little bit more than an 1/8. I not sure what to think of these numbers but I hope this test cut helps in diagnosing the problem with my cuts. The corners of the squares where the bit plunges initially is also a problem area, the machine leaves artifacts that deform the corners. I also don't know why this happens, I changed the plunge rate to a lower number than what the chart says to use but it still does this. As a note, I also changed bits to a Spektra coated 1/4 bit I got in the mail to see if that would yield better results, but no dice :/. When I added Loctite to all the pullies in the past I made sure to tighten the screws well and made sure they were hitting the flat spots on the motor's shanks. But if that's still the problem then it's most likely on the Y-Axis from what the test cuts results say. Maybe I'll flip the orientation of the test to see if I get results that confirm that the Y-Axis is out of wack. Hope this all helps and let me know if you need anything else :]
When you calibrated, how far did you measure? I did the OpenBuilds wizard method in the Control Software, but after getting it as close as possible, I decided to go the old fashion way and sent it 1000mm along a meter stick and then do the math. It was much more accurate that way. Math formula here gnea/grbl. Also, although not your problem here, as Sharmstr mentioned, if you are trying to fit one part into another you need to know the exact diameter of your endmill for creating your tool paths in your CAM software or the parts will not fit properly. Most endmills are not the size they are stated to be.
Later versions of CONTROLs wizard allows you to enter how far you want to go, so no need to work out from the formula yourself (;
I was able to fix the problems with getting inaccurate cuts by using the calibration wizard on, at least with the square on the left. Since I got the square on the left cut accurately I thought I'd try recutting the first test cut (On the right), but still, I got similar results to when I first started. My hypothesis is that the table the machine rest on isn't rigid enough, could this be the reason it preforms poorly with curve items? Or maybe the machine is still out of square. Here you can see that theirs a variation in width on the T shaped part. All the should all be one inch, but they're all diferent, even when they're on the same axis. Again maybe this has to do with the table it sits on, or the squareness of the machine. What I plan to do is move the machine onto the floor and see if it preforms better there, and if it does the machine needs a more rigid table.
I managed to move the machine onto the floor with some help and got all corners of the machine square. After that, it was time to cut the same interlocking pieces that are in the previous post. But got disappointing results, the same problem I had before is still happening even when the machine is on a rigid surface and it's square. I even flipped the way the pieces were cut but got even worst results. I'm at my wit's end at this point, not sure what to try next. Any suggestions would really help.
What works for me when I have to be fairly accurate is to ramp into my profile cuts leaving about .010" to .020" radial stock. Then take a finish pass at full depth to clean up the stock that was left. Lots less stress on the bit. I almost always have to do test cuts of the parts to dial in the dimensions when accuracy matters.
So I tried using a roughing pass then cutting it to the final dimension as suggested and got this, it's one of the better results but still not good :[. I think I'm gonna call it quits, for now. Starting to get tired of cutting the same thing over and over again, plus I need to get more plywood. Let me know of any other tips that could help in any way.
Yeah. I'm guessing either... 1) Grab the collet nut and give the spindle a good wiggle. Feel for what's loose somewhere. 2) You're taking way too heavy cuts. Dial it back to something absurdly small, like 1mm DoC, and see if it's still doing it. That's red oak, it's not a very forgiving ply to cut. It was accurate and square in SYP sheathing above with a shallower DoC, no reason to believe this may not be the issue.
So I discovered that the Y2 motors timing pulley was loose after taking the machine apart, it was loose enough that I was able to pull it off by hand. After getting that pulley locked on the shank with the help of some blue Loctite I put the machine back together and made sure it was all square again. Then made sure the router was trammed up to the spoil board. Afterwards I calibrated the X and Y axis on in OP Control after making my first test cut. Now it was time to cut the real test piece. I used a different plywood then before cause that's what I had on hand, but I ran the the machine at 1mm DoC as suggested. and I got the same kind of cut, I must be doing something wrong. Maybe I didn't tram it well enough. I flipped one of the parts and it fit together a a little better then originally. Which leads me to believe it is a problem with the tram.
