I am using CamBam for GCode, and Chilipeppr with a TinyG...as shown in the pictures, the router did some unexpected things..as you can see, that crazy move out of the F without lifting is not in the drawing or the code...where do I start to look? Which is the likely suspect?
Issue 2 Same setup as above...looks like it skipped a couple of lines on the second cut, and it shaved the top of C down for no apparent reason...I dont think it droppped any steps, because wouldnt that thro everything else off? After it screwed the C up on the first pass, it did the first of the cut okay, then screwed the C up again...
I ran the code through several viewsers, including CutView, everything looks okay, does that mean soemthing going wrong/wrong settings on/in the TinyG?
Doing some more testing, cannot cut a round circle...I have attached the TinyG settings as well as the GCode. Thanks JT, but if it was losing steps wouldnt it screw up from that point on? Just making sure I understand about losing steps...also, everything is tight, nothing moving that shouldnt be... The start point is the right side, it is moving in a straight line before it starts the arc.
I solved the riddle of the circles not being round! After much learning about arcs, looking at all of my TinyG settings, and other head/scratching/hair pulling, it dawned on me that my X axis might be two tight, I loosened it off some and viola, everything (at least with the circles) is beautiful...I dont think this explains my first post, but after I make some more test cuts, I will go back and recreate the file for my first issue.
Easiest way to find out if you have lost steps is simply to check to see if doing a returns to zero after a job actually returns to zero.
I have been having the same problem with my X axis. Don't want to hijack the thread but it's got me stumped. Attached is a photo of an PCB etching I did. It does this on regular circle cutouts as well but not as noticeable. To try and solve my problem, I tightened the crap out of my belt to make sure there is no play. After reading you fix, it looks like that makes it worse? I'm also using TinyG with latest Edge FW.
The arcs to me look to be right, how possitive your pulleys are not able to twist slightly on your motors? I found you can't get enough torque in the standard m3 grub screws to be certain of it, and found they were always loosening. Drilling out and tapping with m4 screw solved the issue.
I'll give it a thorough look see this weekend. It's really odd to me because if you push/pull on the X axis, it's solid with no give. Then again, you probably can't notice that small amount of give anyway.
Also, check the drag on your eccentrics and re-evaluate your speeds and feeds. If the eccentrics are tight and you're cruising along at a high rate then the motors may lose steps because of the reduced torque inherent at higher speeds. Just a thought. Joe
I've had several problems with my OX since I built it. The first one concerned the Z-Axis. I had the speed set too fast and it would lose steps occasionally when going up. I'm not sure what caused it to start and it didn't do it all the time. The downside was that it wouldn't advance high enough to clear the wood and would cut when it was supposed to be traveling. When it was cutting, sometimes it cut through the base board because it was too low. I reduced the steps to increase the torque and reduced the speed. That one is fixed. The second problem also concerned the Z-Axis. I had a dust collection shoe on it. When I went straight down, the bristles caused the router to tilt slightly which played havoc with hole spacing as well as depth of cut. I removed the shoe and I'm still working on a solution. I'm increasing the rigidity of the Z-Axis and redesigning the dust collection system. It works fine without the shoe attached, but I can't take all the dust. The third has to do with the Y-Axis. I'm losing steps and can't figure out when or why, but it will sometimes lose 1/16" for no reason that I can discover. I tried increasing the current to the motors, but I had work to do and I'm not sure the problem is fixed. I'm in a different state from the machine right now and can't check it. I also plan on reducing the steps on that axis, too, if the increased current doesn't fix it. All in all, I've been really satisfied with it. I've spent a lot of time learning to deal with the idiosyncrasies, but it works quite well. I'm constantly amazed it works as well as it does. The bottom line is if you spend a little time with yours, I'm sure you will figure out what it needs.
The same tb6600 chips that I had issues with afraid to say. I can't be certain that current was the only issue that caused them to loose steps but I believe it to be a big factor. Mine were rated 4.5a but even with hefty heatsinks and fans I was only getting 1a repeatable current from them. Swapped to chopper type mosfets cw5045 drivers that do output thier rated current repeatably and not lost a step since.
I'm only having the problem on one axis. Since I reduced the speed on the Z-Axis, no more problems with it. When I was originally having trouble, I tried moving the drivers around and nothing made any difference. Never had a problem with the X. I have the Z set to 1.8a, the X set to 2.2a and the Y set to 3.5. At those currents, my drivers don't even get warm. They are rated at 5a. When I get back up there, I'll put the scope on it and verify the waveform and I'll check the current.
This is spot on. I found that when possible it's easier to use temporary thread locker. Like blue locktite. A problem I'm having right now is that I'm dealing with a solid coupling next to a plastic shaft collar. Using heat when removing set screws may have negative impact. I may have to drill and tap. My only reserve at the moment is the alignment with the screws and tapped holes in the shaft when it's back together. I guess I could use a transfer punch though.
Me too!!!! (pulley set screws)...I spent hours with what seemed like different issues only to find out the tiny pully screws had loosened! I drilled them out to 8-32 and they have been good so far...
I can't remember if I had any concerns doing it as it was a fix or bin situation. Simply pulled pulleys off, removed grub screws, drilled out and retapped for m4, pushed them back on and I didn't even have any m4 grub screws at the time so I used m4 vwheel bolts without cutting down with heads sticking out lol. Ran with those for months.
I ran into a similar issue when using Marlin/Arduino. The firmware was not enabled to go less than 0 axis position. Since the kerf on ESTL CAM starts in negative coordinates, it made circles and some parts chopped off.
