I have a pretty much stock C-Beam machine that I built a while ago but for a number of reasons I haven't been able to work with it. Finally was able to start using it and am having a problem with getting accurate cutting in some cases. I'm using SketchUCam to cut a number of holes in the side of a stock box for some electronics (for an astrophotography tracker, though it's unimportant for this problem). Anyway, you can see in the Sketchup image, the holes I'm trying to cut. And in the photo of the test cut, you can see that I'm having problems cutting square holes (well, square within the confines of the radius of my 1/8" cutter). The initial plunge appears to be off by about 1 mm. These are inside cuts, if that wasn't already obvious. I believe this problem is backlash and tightened the delrin antibacklash thingies on the X and Y axis but that seemed to have no effect. I've read the forum and see a number of things to try but wanted to be sure I'm chasing the right problem. 1mm seems like a lot.Any thoughts or suggestions are most welcome. For what it's worth, I've cut test squares and circles with an outside cut. The squares were cut to 100mm x 100mm and are within 0.1 mm. The corners are clean and precisely 90 degrees. The circle, 80mm diameter, is also within 0.1mm of round. 1/8" endmill and 5mm MFD(ish) stock. I used SketchUCam to generate the G-Code. Running GRBL 1.1f and using GRBL Panel.
Hi @phil from seattle, we need a bit more information - others with more knowledge of your system will probably chip in with their own questions - but to start with what Depth Of Cut, speeds and feeds are you using for those slots. Did that differ from the test squares and circles you did? What controller and stepper drivers are you using? Alex.
Depth of cut is 5.2mm. Material is MDF. Controller is an arduino Mega running GRBL 1.1f. Stepper motors are the standard ones that ship with C-Beam package. I believe they are these Drivers, I'll have to check exact numbers when I get home but they are separate "high power" drivers (i.e not step sticks).. Feed rate was 1000 mm/min though I'm not 100% sure about the earlier tests. Router is a DeWalt 611 running at the highest speed. Bit is a 1/8" carbide upcut router bit.. I've got some family obligations for the next 2 days but have cooked up some hand written G-Code that will test the backlash theory at different feed speeds to put that issue to bed. Probably won't get to it until sunday.
turn on ramping at 10 degrees. router bits do not drill straight down with any joy. turn on multipass and do 2 or three passes. this relieves tool pressure. but most importantly I believe your Z wheels are loose and Z is tilting from side to side from the tool pressure.
As @David the swarfer said that is a hefty DOC for a 1/8" router bit - even single flute bits where there is some cutting edge in the centre won't like drilling 5.2 mm in one pass, and then you are asking it to cut a slot that depth the full 1/8" wide. That's a lot of sideways force on the bit. Alex.
So, I had some time today to take a look. The z wheel assembly was not loose - wouldn't move when I push on it - though it is at the limit of the eccentrics. I did find some play in the X assembly - the collar on the lead screw had a small amount of play. Unfortunately one of the left side retaining block bolt's hex socket is stripped - bolt seems seized. Had to loosen the other end by the stepper. Tightened the collar up. Also tried multipass - 2mm/pass and ramping. The results were better though not perfect. Good enough for simple stuff. I will experiment with even shallower cuts. There appears to be some gantry flex when I push on the Z motor. It looks like the rail for the X axis is flexing just a little. Not sure that's fixable without moving to a beefier rail and maybe uprights - seems inherent in the C-Beam design. I think there is little hope for cutting aluminum on this machine. Anyway, thanks for your help.
the c-beam machine was designed for cutting aluminum so if yours will not do it then something needs fixing. did you follow the instructions?
Yes, I watched the video several times before starting and then assembled with the video running (more paused than not). Did a lot of squaring and checked my machine for square yesterday - It's still square. Also, I tightened every bolt/screw yesterday - though most didn't need any. Have to admit I'm worried about over tightening as I have one that has the socket rounded over (if that's the correct term) - not sure how to get it out but that's a problem for another day. How do I deal with X axis beam flex that causes Z axis motion? Just do shaving cuts? Seems like I can do MDF with 2mm passes. What does it take to do aluminum on the C-Beam? I'd love to do 1/4" plate. Is there a tutorial on that somewhere? I noticed a post from someone that strengthened his C-Beam machine with steel - especially the X Axis beam. Maybe that's worth doing.
So, a quick update. I went to multipass 1.5mm and 10 deg ramping for my 6mm MDF and am getting better results. Also trying a different endmill that seems to plunge a bit more aggressively - these - with much better results. (this is, I think, a very important point - good/right bits make a big difference.) I also went back and rechecked the machine based on the original assembly video. Didn't make any changes though, it's pretty much in "spec". I think it is pretty solid though the Y axis varies in Z about 0.5 mm front to back. I could shim the back end of the cbeam but opted to surface the wasteboard with what looks like perfect results. I've cut about 20 things in MDF and BB with good, if slow, results. Still haven't tried aluminum yet - lots of projects, not enough time. I'm getting circles that are accurate to about 0.1mm in 70mm which I think is OK for wood and plastic. Not sure how accurate my measurements are on even a lightly sanded piece though but that's a different topic. The upshot is that the C-Beam, when used within its limitations, works pretty well. I need to experiment with the feed rate to see if I keep it or move to a more capable machine.
