"Firmware unresponsive..."

[CTWalbridge]

Hey all - I'm having a problem with my M2 after a month of successful prints: During the print process, I get a message in S3D's Communication window that says "Firmware unresponsive. Attempting to force continue..." and the print head keeps going outside the perimeter of the print, I think until S3D reestablishes communication. Or the print head moves without extruding.
- It has only happened with one group if STLs I've wanted to print, some Spirograph gears.
- I made sure the STLs I downloaded were manifold by running them through a Netfabb repair, a step I had skipped the first time this happened.
- I've made the GCode with two separate recipes, one of which has been successful in the past.
- I reprinted something that had been successful in the past, and that worked just fine, then I had the same issue with the Spirograph prints again.
- The variable I haven't controlled for yet is that my successful print was from the SD card, and all the others were over USB, so I'll try that next.
- I'm running an old Mac with OS 10.7, and that's as far as that computer will ever be updated.
Any suggestions? And how do you update the firmware on the M2, and do you ever need to?

[jsc]

So, one thing I've read about as a potential cause for missed communications when printing through USB is an electrical phenomenon called "ground bounce" or "ground loop". Perhaps others more knowledgeable about this could chime in, but one surefire solution is to always print through the SD card. Another approach I've seen is to drive the printer through a Raspberry Pi with the ground wired directly to the printer board's ground. I print through a Beaglebone Black plugged into the same outlet as the printer and through a short length of USB cable, and I haven't had any missed communication problems, although I'm not sure if that addresses the ground issues, or I've just been lucky.

[CTWalbridge]

Today I printed the same GCode via the SD instead of over USB, and it worked. I'll continue to test - I've had success printing over USB in the past - but thanks for that advice.

[Tim]

The "knowledgable response" is that USB uses differential signaling, so it can't really suffer from ground bounce, although a large ground difference can reset the RAMBo on some earlier board revisions. There are many possibilities for what the problem could be, such as a baud rate that's set too high, or a latency that's longer than the timeout period, etc. All of these problems would be down in the USB driver, and whatever they are, you can avoid them all by running off of the SD card.

Although you won't be upgrading the Mac, there's always the possibility of updating the USB-serial driver.

A likely relevant topic is: http://www.forum.simplify3d.com/viewtopic.php?t=794