I know this is more of a question for the software vendor but thought I would post it here as well to see if anyone has had similar issues. We just got our first build (c-beam plate maker) completed and it seems to be working almost perfectly. We cut the 'hello world' test file with Chilipeppr and it worked great. After that, we designed several letters in Vcarve Desktop. We tried cutting out one of the letters and the g-code did not seem to translate correctly to the arduino. We have the arduino set up with the Gshield. I selected the generic gcode post processor (inches) in Vcarve. The material we are cutting through is .25" thick. We set it up for two passes with each pass cutting .125". The first pass on the inside of the "R" that we were trying to cut worked fine but the second pass just mirrored the first pass without lowering the Z axis to cut the second .125". After that, it moved to the next part of the tool path and the Z axis actually raised up about 1" and just traced around where it should be cutting for the first pass and then lowered for the second pass and cut the first '125". Do I need to use a different post processor or is it not a post processor issue? Thanks in advance for any input! Nate
I was able to determine at least part of the problem. The Z axis was bottoming out. The test cut that we initially made was not a through cut, so it did not appear to be set wrong. I adjusted the spindle and reran the cut and now it is cutting too deep. At least there was some kind of progress though. =)
The only thing I can think to check is to make sure that you have your units (mm/inch) the same in your design as your post process. After that make sure to check your steps per unit on your card and also check to see if your Z axis steps mode jumper is maybe in a different stepping mode or something simple like that.
Here are the settings that I have on the board as well as what is showing up on chilipeppr. By process of elimination, I believe that it has to be something on the board, I am just not sure what it is. The CAD and the CAM software both show the proper operation as far as the steps go. I ran a simulation on both Vcarve as well as Chilipeppr and both show the proper passes. Nate
It might be that the Z-axis motor is losing steps. What is the voltage of your CPU and how much Amperes are you giving your motors? -Ronald
Here are pics of the setting for the ampere settings. I have not messed with these at all. I guess this is something else I will need to look into.
So after doing some more reading, it looks like I might want to reduce the Z settings on the board. They are currently 200 step/mm and 2000 max rate, mm/min. Would cutting both of those by 50% be a good starting point? I guess I should add that I have NEMA 23 motors as I know that plays into the amp settings.
Hold on there Nate, Those settings have nothing to do with your problem unless you move an axis 1mm in your program and 2mm on your machine. 2000mm/min max rate is the best for top speed for your NEMA23 motors and Acme screw, go faster and you get resonance. When you use a CAM program your cutting speed will be less than 2000mm/min unless you cut something real soft like real soft foam. Set the right amperage for the motor drivers first. Let me see if I can get you a good link for the gShield. You only need to know how to use a volt-meter . -Ronald
https://github.com/synthetos/grblShield/wiki/Using-grblShield is what you should read but it's a bit... well, "turn a notch back when it fails" explanation... there must be something better...
Thanks for the link! I think I have read through it before but I have not done the test that it outlines for establishing the limits on the motors. I guess I will give that a shot tomorrow. Nate
You can't go off the pots with these type boards, They are rated on peak amps and not repeatable amps so will be underpowering the motors if you have set at the motor rated amperage, crank the nema 23 channels to max and it will he outputting around 2.3a per channel and the nema 17 to 4 amps which is actually around 1.7-1.8a. If you are concerned, even if the amperage was true the motors would still be fine for a good while and you will notice you can barely touch them before they will do any damage to themselves. they will just get hot quite quickly, so just feel the motors from time to time, if they are too hot to hold they are over powered, but correctly powered they should be on the warm to hot side. Generally they can go upto 110celcius without any problems. My 4.2a rated nema 23's powered at true 3.8a run at around 50-70 Celsius to give you a better idea.
I just wanted to report back that that we figured out what the problem was. We went through the amp setting process and were able to set the low amp setting but even turning the pots all the way up, the motors would not stall out. I started reading on that link about microstepping and a little lightbulb went off. The Gshield that we have came with only one jumper and it just so happened to be on the Z axis. This meant that the Z axis was running on 2x micro steps while x and y were running at 8x microsteps. I removed the jumper and everything is now working perfectly. I know this is a rookie mistake that I should have caught earlier. I do appreciate all the help though. Going back to the amp settings, due to the fact that turning the pots all the way up does not make the motors start turning off and on, I guess I should just do the touch test as I am cutting? Thanks again for all the help! Nate
Right. Sorry I missed that jumper on the photo else I would have told you to change it. Odd thing is, there was only one jumper with the board while you might need 6. -Ronald
Yeah, that is kind of what I thought after understanding how the jumpers worked. My simple mind thought when first looking at it, that I only needed one jumper to set the microsteps for all the axis'. In the end it was a good learning experience. Nate
Yes, it's really the best way without measuring the amps but not all of us have a couple of thousands worth of probes laying around to do so accurately or a Btec in electronic engineering to know how to do it mathematicaly, and yes there will be a few shaking their heads and shouting manufacturer specifications at me but it's as good a test as any and certainly more accurate than the pots, I personally know how hot my motors should be and it would ring bells if after half hour of cutting they were cool. Or if they were to hot. I can pretty much be certain that board will run 2.5a nema 23's at max current and will be just about right at 2.3a. Sound is also a good indication, with the pots up max for nema 23's, jog around, tune your ears to the motors, now back them all off by 1a and jog around. Do they get quieter or louder? If it gets louder, with this board it means at max current you are Closer to powering the motor correctly and if it gets quieter which it may do then chances are even at max amps you are @ less than 50% of the motors rated amps and it's getting quieter simply because it's like a motor running out of batteries. If it's overpowered it will sound lovely and quiet, just get hot. Basically Around 90-100 Celsius is good time to consider dropping the amps.
