Stops at random points in the middle of the print

[Jules]

Here's something interesting. I disabled all the Windows power management for the USB port device drivers and replaced the USB cable. Been printing for 3 hours with no pauses. Hum...

That IS a problem with some cutter software as well.

[Tim]

i have noticed mine having a slightly scanty startup on new layers, but that was deliberately introduced (by me) with a negative flow to try to offset that globbing business that now shows up. After a while, mine generally starts extruding again....but it's making a bit of a mess - particularly on that skirt and first layer - they're not nearly as neat as they used to be. (I'll be working to fine tune it, but it would help if I knew what was causing the blob.)

I think this is becoming a topic for another thread. . . I have additional comments about using the $#@! prime pillar.

Did they update S3D v3.02 a while back?

Yes. The files in the distribution are dated August 17, 2015. I downloaded it on October 6 as part of trying to figure out if they had fixed any of the dual extrusion problems.

[dklassen]

Been printing flawless for the last three weeks, couple dozen prints at least. Out of the blue the pause problem is back. I don't get it.

[Tim]

Been printing flawless for the last three weeks, couple dozen prints at least. Out of the blue the pause problem is back. I don't get it.

Well, I'm still confused as to whether I'm experiencing the same thing or a completely unrelated problem that just happens to do the same thing. Occam's Razor says it's the former, so I'm going with that.

Have you eliminated any possibilities? Things that you started or stopped doing since you first started having the problem?

Unfortunately, like all intermittant problems, it's easy to fall into the trap of making some change, having the problem disappear, and therefore assuming those two things were causally related. . .

[Jules]

Been printing flawless for the last three weeks, couple dozen prints at least. Out of the blue the pause problem is back. I don't get it.

Did you check this again? We used to have the problem recur with some of the background Windows updates.

I disabled all the Windows power management for the USB port device drivers

[dklassen]

I've checked and rechecked everything I can find and read just about everything I can find related. Started to go the OctoPrint route but can't even get the Wifi going. Spent about 7 hours so far just on that. I'm ready to give up and just spend the money on another Mac Mini to run the printer and be done with it. Crap sakes...

[PcS]

I will say after you edit the file for wifi. It took my pi 2 two reboots and then had to reboot the router for it to go. But other than than all pi wifi setup has been easy for me. I know some others have had issues as well though.

[ednisley]

spend the money on another Mac Mini to run the printer

Makes Pronterface and Mint Linux running on an old laptop (*) next to my M2 look downright attractive. That setup doesn't stall while printing over USB, doesn't randomly disconnect while printing, doesn't nag about upgrades, doesn't mysteriously lock up, doesn't get in the way. It Just Works and has done exactly that since I set it up a few years ago.

It's on the house WiFi network, so I do the solid modeling & slicing from The Comfy Chair, trot downstairs, load the G-Code file from the shared directory tree, fire the M2, and It Just Works. Copy the G-Code files (actually, the whole tree) to the laptop's SSD, carry it to a demo with the M2, and It Still Just Works without a network connection.

At show-n-tell demos, it runs OpenSCAD and Slic3r, too, with no licensing quibbles. Granted, it's not the fastest machine in town, but entirely adequate for the purpose.

I mean, sheesh, I know everybody wants to make this whole 3D printer control thing more difficult, but why bother?

[/rant]

(*) Under $200 from Dell's off-lease outlet, many years ago. Nothing fancy. Remember that a $35 Raspberry Pi costs about that much after you add a power supply, case, WiFi adapter, LCD panel, cute keyboard, and blah blah blah.

[PcS]

The pi3 being 35 dollars with built in wifi is a no brainer. Most tech people have a spare sd card and or web cam laying around. It can be setup on your existing monitor and keyboard mouse combo and then ran totally through the browser on a network. All in all it is much easier than any other option. IMHO.

[Tim]

but then I tried printing using SD card rather than over USB and it's still doing the same thing.

This part of the original post still bothers me. Mostly because I got similar behavior; I transferred my print to the SD card and it still stopped dead in the middle of the print. I was inclined to conclude that the USB interface isn't implicated in the problem. But then, as long as I have S3D running, there's still a communication channel open on the USB, and S3D is making periodic queries about temperature and head position. There is nothing in the G-code itself. I can track down the point of failure and there's nothing there except normal position codes.

I can only think of one thing that fits the evidence, although it seems unlikely, which is that S3D is sending a pause command to the printer in the middle of a print. My reasons for suggesting this are that it appears that the problem is happening only to people who are connecting to the printer through S3D (regardless of whether they are printing over USB or off of the SD card), that the problem appears to be OS independent (I'm using Linux), and firmware independent (I'm using Smoothie), and that the problem acts very much like a print pause, in that if I hit "Pause/Resume" in S3D, the print will resume (I have to hit the button twice, since the S3D GUI is unaware that the print has already paused).

I guess one thing I could do is to add some code to the firmware that would write to a log file the time of any pause or reset or any other unusual command that arrives in the middle of a print. I spend much of my day job tracking down software/firmware/hardware bugs, so it's something I'm supposed to know how to do. . .