S03 suddenly went rogue

[sjj47]

I was milling a dustshoe prototype when my PC crashed -- the old blue screen of death. After rebooting, I reran the job using bCNC, and after two minutes the bit suddenly drilled down and X-Y movement stopped. No error message on the bCNC.

I rebooted the PC again, restarted the job, only to have the same thing happen at a different point in the nc file.

I rebooted a third time and tried using UGS instead of bCNC. This time the job was 90% done and suddenly the bit veered and dove down again.

s03_rogue4.jpg (199.37 KiB) Viewed 1052 times

s03_rogue3.jpg (195.18 KiB) Viewed 1052 times

The file looks fine in OpenSCAM simulation. Also the problems occurred at 3 different points in the nc file, so maybe the cause of the problem is not the nc file itself? Since the problem occurs on two different gcode senders, maybe it is in in the PC - a java update with a glitch? I know after the crash the PC spent a lot of time installing updates to various apps, so maybe one of those updates upset the apple cart?

To test that, I tried running a big job (+40 minutes) that I had run successfully before. I figured if it failed now, it would because something in the PC had changed. But it ran without trouble. Now I'm back to thinking it is a problem with the nc file, although why it should break at three different points I don't understand.

Anybody have any thoughts?

Thanks,

[WillAdams]

Usually the computer stuff is binary, either it works or no. My suspicion would be that if it is the computer, it’s going to be maddeningly difficult to troubleshoot, and it’s some variation of the grounding disconnect: http://www.shapeoko.com/wiki/index.php/ ... mendations

More general troubleshooting stuff here: http://www.shapeoko.com/wiki/index.php/ ... rmittently

Double check the belts and pulleys, but it’s likely electrical.

Contact support@carbide3d.com if you can’t puzzle it out.

[sjj47]

Thanks, Will!

Your response got me thinking. I had been vacuuming up the dust with hose connected to the shop-vac -- which terminated in a metal tube. Could the flow of air and dust have created a static charge? And then grounded onto the S03 chassis when I inadvertently touched the tube to the router block?

It sounded possible, so I reran the job taking care to avoid that situation. The job completed successfully. So have a couple of other jobs. Not dispositive but indicative.

Thanks!

[WillAdams]

Yes, electrostatic discharge from a vacuum is a potential issue.

[madhatter]

Thanks, Will!

Your response got me thinking. I had been vacuuming up the dust with hose connected to the shop-vac -- which terminated in a metal tube. Could the flow of air and dust have created a static charge? And then grounded onto the S03 chassis when I inadvertently touched the tube to the router block?

It sounded possible, so I reran the job taking care to avoid that situation. The job completed successfully. So have a couple of other jobs. Not dispositive but indicative.

Thanks!

It definitely can cause your SO to have issues. I solved my "SO3 disconnects when I turn on or off my vacuum" issue by hooking up my SO3 through an APC UPS. I had a second issue, and that was when using the vacuum, it would occasionally disconnect during a job where I Was cutting wood, and it happened every time I cut plastic. I bought an anti-static vacuum hose and grounded it. I didn't want to fork out the money for a metal nozzle, so I wrapped it in aluminum tape and grounded it to the hose, and since then I have not had a disconnect from static.

In fact, I have only had one disconnect in the last 6 months. It was because I disconnected the hose ground strap to suck up some sawdust because I did not have the dust shoe adjusted properly. 30 seconds after disconnecting the ground strap, my SO3 reset. I was not a happy camper when I realized what I had done because I was too impatient to wait until the job was done before vacuuming up the sawdust.