Machsupport Forum
Mach Discussion => Mach4 General Discussion => Topic started by: JoelyB on January 31, 2021, 12:03:06 AM
-
Good Afternoon,
I am trying to run a boring operation in Mach 4 using G Code that I have run previously (a week ago) with no issues. For some reason when running the same code again today it keeps on behaving in a strange manner and starts following tool paths that are not in the gcode and only appear in the tool path window once it commences the helical move for the boring op. What on earth is occurring here?
TIA Joel
-
Please look at the history line and tell me if you see anything. Could be that you have a softlimits issue.
-
Nope. Nothing there. Just the setting of the tool height.
-
And it has done it again, and ruined a part this time!
It got up to the boring op and then added a random toolpath, instead of completing the boring operation.
I regenerated the tool path from fusion 360 and then ran it in mach 4. For some random reason at this point, the machine coordinates shifted 90mm in the Z axis, 5mm off causing a crash.
I have attached 2 images. The first is the toolpath with the additional toolpath that does not exist. The second is the same g code just regenerated. What is happening here?
-
Can you upload the gcode file? I will take a look at the code. Also by any chance you didn’t happen accidentally press the y axis zero button when changing the tool or something. Maybe we need to lock out the zeroing of the axis. I have seen people lock out the zeroing so this can’t happen. We will get to the bottom of this. Somehow you have added to your fixture offsets.
-
The setup with 3 ops, was the gcode that I was running when this occurred. When I opened the second file to run it on its own, that is when the machine coordinates for z increased approx 90mm.
It is worth noting that I when I first had this issue I was running 4300. I am now running 4580 and it has occurred again. General user experience seems to be better with this version.
-
This is what happened...
-
The G-code looks okay to me, I am not a fan of G28's but your using them correctly.
That is some funky Gcode! I sort of like it. I had to do a double take when I saw the plane change in the code ;)
So your correct it looks like the issue is at the toolchange. You by any chance don't have an offset for the Y axis setup in the tool table for tool number 7? I know it is crazy.. Also I may need to see how your doing your toolchange and setting the Z axis offset for the next tool. Sorry I can't help more now but we will get this! Please run in air so we can make sure we have a process that works.
Thanks
Brian
-
Hi Brian,
A few more details to this issue:
The fault occurred in the second boring op (BORE 2 2).
I am running a MB3 breakout board with an ESS.
As the first occurrence of this was also with a boring op (different part), so this op seems to be the issue.
If you look at the image in the first post of this thread, you can see that y has travelled to y-128.1554 where as the Gcode instructs it to go to y-30.385 and the line before was at -28.385.
I can't see how it could be the GCode. Any thoughts on where to go from here? Is there any testing I can do to help troubleshoot?
-
I am thinking it could be an accidental press of the zero button at this time. Could you please try to see if it will do it again . I don't really have anything to go by at this time. Also make sure you don't have any X or Y tool offsets .
-
That would make sense except for the tool path that doesn't exist in the g code that can be seen in both occasions that this has occurred.
-
Okay I get it, I just need you to test for me if you would please . if you can make it happen without making a part I can then get your profile from you and test it here. I want to fix this but I can 100% tell you that I am not having this on my system. There are many ways to make this happen and I just want to make sure we are looking for the correct one. We will get this.
-
Sure, thanks. I will give it a shot tonight
-
Do you have any code in your M6 macro?
-
No, I don't have any code in the m6 macro. I have 4 PMC scripts. When this issue started I reinstalled mach 4 and setup a new profile and reconfigured everything to make sure it wasn't my profile causing it. For some reason a couple of the PMC scripts no longer work. (Not sure if this is relevant)
1. Turns the optional stop on as soon as the system is enabled. (Working)
2. Feed hold button (Not Working)
3. Cycle start button (Not working)
4. I can't remember the fourth but will edit once I get to my computer tonight.
-
Do you have a tool changer or a tool setter?
-
No, I don't have any code in the m6 macro. I have 4 PMC scripts. When this issue started I reinstalled mach 4 and setup a new profile and reconfigured everything to make sure it wasn't my profile causing it. For some reason a couple of the PMC scripts no longer work. (Not sure if this is relevant)
1. Turns the optional stop on as soon as the system is enabled. (Working)
2. Feed hold button (Not Working)
3. Cycle start button (Not working)
4. I can't remember the fourth but will edit once I get to my computer tonight.
Correction. None of these scripts were active during the issue
-
Do you have a tool changer or a tool setter?
No. All manual.
-
I have some gcode to bore 4 holes. I will run it tonight in air and see if I can replicate the issue.
-
Hi all,
I ran the 4 boring operations gcode (see attached - Boring op test part). It was supposed to bore 4 holes. Initial generation of the tool path looks great. As soon as it started to move to the first hole location, the tool path window changes and there is this new magical tool path that does not exist in the code which it follows(See attached Boring four Holes test 1&2).
I then reset everything and after trying a few things I cannot get it to repeat the same behavior. Got to love intermittent problems
-
The only issue I can see in the tool path that occurs more regularly but not consistently is the image attached. *The line should be straight but it has some strange triangle mid path.
-
Ok, I have attached a log with the triangles occurring in the toolpath at the end of the log. I ran different gcode first and tried to reset things as much a possible. I hope this helps. I have also attached a screenshot.
-
I think we need to know a bit more about your machine. First, do you have encoder feedback? What are you running for electronics. It seems to me like we have a system that is trying to sync with with something that is not in the correct location...
I did look at your posts here and I didn't see a "parts" list for your machine. Could you please tell us a little bit of what your setup is?
Thanks
Brian
-
Not sure it will help me but I would like to look at a packaged profile.. AND get a Dev version of you are not on the latest code. We had worked on closed loop code and Aux feedback in the latest versions and that may help you out.
-
Parts list:
I have a converted Optimum BF30 Vario mill converted to 3 axis CNC.
3x Nema 34 12Nm closed loop stepper motor's from steppersonline.
3x 60v power supplies
1x 24v power supply
MB 3 breakout board
ESS
Spindle control via 0-10V Anaspeed circuit on mb3. Using the stock optimum spindle controller.
I think that should be it. Did you want me to add some of my config files?
-
Did you connect the encoders from the steppers back to the ESS ?
-
No just to the motor controller. My understanding is that I would need some sort of signal duplicator to do that?
-
Something very odd is going on and I can't get my finger on it now but ... we will!
Any chance you can tell me what you have for a PC? I wonder if we have some power saving issues .
-
An Acer laptop with Windows 10. The laptop is plugged in and the fault was occurring approx 5 s into clicking enable, so it is unlikely going into power saving mode so quickly after being active.
-
I have been thinking about this and I would like to have Trevor log in and check your version of Mach4 and ESS plugin. He can also check the power settings of THC Ethernet port. When would be a good time to have him remote in?
-
Sure. I could do 0700h AEST on Saturday here. Which would make it something like 1600h Friday in Maine?
-
Okay I told him about it and he is ready for you ... He thinks he knows what to do so we have hope :)
-
I hope you guys will post the results here for everyone else to see. I’m not having this problem, but it’s kinda like listening in on a party line and I’m courteous how it all works out.
Thanks
Art
-
We sure will :) I want to know myself what's up!
-
Any progress on identifying this fault Brian?
-
Mach 4 gave me a good one today.... Thankfully I was checking the toolpath screen carefully. In the first attachment you can see that the tool path has nearly duplicated itself but approx -50mm in z >:( Which would have driven the endmill into the bed. After a regen of the toolpath all is well again. (See attachment 2)
The GCode is attached at 3.
-
It looks like you have something shifting your set point. Like re zeroing your axis would do.