Hey guys first post on here. I just wanted to say thank you for the wealth of knowledge available on cnczone!!! I just finished up my second build the first one was converting a g0704 to ball screws and milling out all the standoffs and motor mounts. The second build is the pro4848 from cnc routerparts.com. First thing I have to say is the design is slick and seems rugged. I was able to cut acrylic no problem to make a dust boot for it. One minor issue I'm having and is when I plane a surface randomly the x axis will bind for just a fraction of a second and it causes the gantry to flex by about .004 and it will leave a dot in the wood from the router bit. I was curious if you guys had any input as what to look for. I am running nema 34 motors from kelininc vs buying there package. I was happy with the motors on my mill so I went with them again. Also I looked at the R&P and it looks fine. As I was facing off the spoil board you can see where it was basically hiccuped and made a dot. All the dots are in different locations. My motors are setup for 1200ipm rapids and but when I was running the job I was making a .0625 deep cut at 100-200 ipm if I remember correctly. If anyone has see ln this or has advice please let me know!!
You may want to check to see that your rails are completely parallel and well-seated in the V-Con clamps. The hiccup could also be electrical or PC-related. Feel free to contact us through our support email or give us a call if you'd like additional help.
When I assembled it I made sure to use a rubber mallet and that the rails were seated properly in the v channel. I am able to slide them up and down the rails without it binding. I can run it using the arrows on the keyboard back and forth without any issues. My motors are tuned so my rapids are 1200 and I adjusted the acceleration so it doesn't bang back and forth. I can't remember the number off the top of my head. It is in random spots it hiccups not the same spot every time. I'll look over the rails again and verify that they are square to each other.
If you can jog it without issues, I'd suspect the issue is probably not mechanical in nature. There might be some issues with electrical noise from your router -- you may want to make sure the router cable is separated physically from your motor cables as much as possible to avoid EMI interference with your electronics. It might be worth running your program in air (not actually cutting) to see if you still see hiccups, and if so, if they persist by running the same program with the router off.
How are your electronics wired?
Did you put everything in one case?
From what I hear putting the drivers, psu, and your computer right next together can be problematic.
The computer is 10'-15' away. The motor drives are all mounted on a board about 6" up and down from each other and 2" apart so I can get plenty of airflow across the heat sync's. The power supply is also next to the drives about 4-5" from them. All the motors I wired up as bipolar parallel and I am not using micro stepping.
I'm going to face off the spoil board again just to test it again and see if I can see what's causing it... I'd hate to be working on a big sign and have it put divots into the face of it so I want to get this thing 100%. I have ran a few small jobs on it and a few 3d parts and they came out great! I've been tuning under 100ipm for them
I was having an issues with missed steps caused because the pulse length I was using was too short. On repeated moves back and forth the machine would drop a step or 2 now and then. I set mach3 to Sherline1/2 mode and pulse length to 15us and now I have no issues with dropped steps on my joescnc2006.
I double checked it and I was already at 5 usec. I'm not however 1/2 sherline. On my mill I was running 0 and 0 on the settings in motor tuning. I changed the settings and ran a small 12x18 piece today and it came out perfect. I'm going to face off the spoil board tomorrow a second time to see if that made any differences on 4'x4' pieces
The longer it is the less pulses you can have in a second (limiting max speed) but if it is too short the stepper driver won't detect the pulses reliably / at all.
I'm rather surprised 0 usec is working, as it means no pulses. I suspect Mach3 is not really sending 0usec pulses...
It was working but not well. I pulled up my spec sheet on the drives and it said that I should be over 1.2. I changed it to 2&2 and ran a 3d part last night and it came out well. I think what was happening was I was trying to run it to fast also ... Guess simple stuff I can get away with a bit faster vs engraving