CNC12 "Sticking" during Macros

[ShawnM]

Look at your error log, starting yesterday it's a wonder it did anything. It's riddled with communication errors. One after another.

Shawn, what file do you look for in a report to see these types of errors?

Scott, when you open the zip file that someone posts there's a second zip file in the report called urf.zip, if you double click that file it opens that up and there's a file called msg_log.txt, double click that file and you'll see the machine history and errors, if any.

[grossmsj]

Great!
Thanks Shawn.

[cnckeith]

you can see these same errors inside of CNC12 in the Utility menu under LOGS

[Ziggy]

As a follow up:

I called Geckodrive today and told them of the issue. They suggested that it might have been a couple of things:

The first is that I might have been taking too heavy of a cut and over-worked one my stepper motors. This seems unlikely since I performed the same cut in the same material later. (See below) Also, it wasn't long before I restarted CNC12 and everything worked again.

The second is that there might have been a drop in my line voltage. I do live in an old neighborhood and do have problems with power. I was in the shop at the time this happened and I didn't hear the vacuum stutter or see the lights flicker. However, I have fluorescent lights. If the power were only to dip I might not detect the difference in either the lights or the noise of the vacuum... especially if I wasn't watching for it.

So I guess the only thing to do is just keep trying.

This leads me back to my original premise when I started this thread. Macros Sticking

I didn't know to look at the logs and they were, indeed, riddled with communication errors. Since those errors only happened on that one day and have not shown up since I suspect Microsoft is to blame. I have since disabled my WIFI. As I would like to get this computer back on my LAN, I am thinking about merely blocking that computer from ports 80 and 443 in my router. I think this should prevent it from reaching the internet. Hopefully Microsoft will not be able to pull updates. (Maybe I can just disable the gateway on the computer.) Does anyone have any thoughts on whether this would be sufficient to "neuter" Microsoft?

Also, if someone could chime in on the time it takes for a Zero macro to run. This is my first experience with CNC12 so I don't know if it is normal for it to take a few seconds for this macro to run. I have written some fairly lengthy batch files / powershell scripts. These pretty much run instantly. Maybe the "Zero" macro reads data for all axis before asking for which axis to zero? Still, should this take this long? The logs do not show any communication errors.

Finally: As I mentioned above, I ran the same NC file that was running when I received the drive fault and it worked fine. I might mention that it was "almost" fine. This was one of those times that the bit plunged an extra .25". So the bit was actually hogging out close to a half of an inch of material before I noticed it. Again, maybe I messed up. But having assumed that I made this mistake the previous week I was being fairly careful about making sure the Z axis was zeroed where I wanted it to be. Maybe I will manually zero it with the "paper" method for a while.

Sorry for the long post. Brevity isn't my forte.

As always, thanks for all the help.

[Ziggy]

I may have discovered something I was doing wrong:

I found this YT video:

She was using the Part Setup menu for all three axis. I was using the button in the attached picture. Is this button just a shortcut for the same thing?

[cnckeith]

I may have discovered something I was doing wrong:

I found this YT video:

She was using the Part Setup menu for all three axis. I was using the button in the attached picture. Is this button just a shortcut for the same thing?

Hi you are using a really old version of CNC12 i(v4.64) free support for that version has ended.

please make screen shots of the v4.64 Wizard setup menus and follow the CNC12 installation/upgrade instructions and install the latest version v5.08

free download and instructions are here.
https://www.centroidcnc.com/centroid_di ... loads.html

[Ziggy]

Sorry for the confusion. This is just a screen capture from the YT video to show which button I am clicking for the zero process.

I think I am one version behind. I installed it last year.

[cnckeith]

there are a bunch of ways to set a WCS zero position.

- use the CNC12 part setup menus. F1 Setup, F1 Part and use any of these menus to set a part zero location

- Press MDI type in G92X0Y0 to set X0Y0 at the current position

- Press a VCP button. VCP buttons are programmable so you can use them as is or roll your own.
https://www.centroidcnc.com/centroid_di ... amming.pdf
https://www.centroidcnc.com/centroid_di ... manual.pdf


they all do effectively the same thing

[Ziggy]

Ok.... I am back.

I figured out what was happening with my machine losing zero. I am sure it will make those with more experience roll their eyes but maybe this will help someone else.

After getting my machine set up with the Acorn conversion, I set up soft limits as I understood them in the manual. They worked perfectly after testing them.

After this I decided to add limit switches so that I could do auto homing and auto squaring. These also worked perfectly.

Since then I have been just trying to get a few projects under my belt to get to know the software. I had noticed that I was able to go past the soft limits and accidentally hit the end stops on occasion but I never investigated as to why. I was engrossed in learning how to use the machine. I should have looked then but I didn't.

So as I stated earlier, there have been times when I set my zero only to have the bit plunge about .25" inches deeper than it should have. When I stopped the carve and used the MDI to tell Z to go to zero, sure enough it was about .25" lower than where I had initially set it.

So this weekend I was being really careful and checking zero from the MDI at every step. Then on the first carve it did the same thing. I didn't know if it was changing when I started a NC file or what.

Then it occurred to me. This happens only after raising the spindle to add my dust boot. On occasion I would raise it too quickly and it would hit the end stop for a split second before I let off. Sure enough. I did a quick test and this causes it to lose track of its Z zero point. It seems kind of obvious now.

So at this point I went into the utility and checked my soft limits. They were all gone. The only thing I can figure is that they were wiped out when I set up the auto homing/squaring.

So this brings up another question: If I set one end of the XYZ limit to the zero position, will the machine be able to go through the auto homing process? From what I understand, it has to go past 0 to trip the switch and then move back. Will it still be able to do this if I set one end of the soft limits to 0?

[cnckeith]

thanks for reporting back!

soft limits are 'wiped out' when you interrupt(cycle cancel or estop) the auto square macro while it is running and it has the soft limits zero'd so it can move beyond the limits to find and square things. it is a well documented gotcha and is one of the cons against the pairing method we choose to use for Acorn. eliminating this gotcha is on our to do list (and it requires a new approach for Acorn axis pairing and auto squaring)