Movement just stops mid execution

[scubanimal]

Good Day All,
Twice this weekend, a long job just stopped after about 10 hours of working fine. I have a very long job doing multiple finish passes (very light) on a delicate 3D image. The point it stops not consistent , the Shark just stops the movement. The effect is like hitting the pause button, the router continues to run, but no movement of the gantry. The odd thing, is that the instructions that typically scroll across the screen complete. I thought maybe my computer times out or other such things, but eliminated those. I saw this once before several months ago, so its not consistent.

Any thoughts?

Thank you,
ian

[4DThinker]

Randomness and ten hours before it stops makes your problem a pickle to figure out. Was the pause recoverable? Had the PC frozen up needing a reboot? Is the PC running the Shark connected to the internet?

I know Windows updates will cause a system to reboot when unattended, and a windows reboot will surely stop the controller in action.
If such a large file, the SD card being used in the Shark may eventually cross a bad sector/chip that it rarely sees doing smaller files. Take out and reformat the SD card may help.

Check the size of the .TAP file. If larger then 2gb it may simply stop copying to the SD card once that size is met.

These are all I can suggest to check at the moment. I haven't done much 3D cutting. Certainly no 10+ hour jobs. My Shark has had its random job-stop moments, but in my case they were likely due to static shock or a power surge.

[scubanimal]

4D, Good ideas;
The job size is about 41K so way under the limit. I do not have that computer on the net, as I don't have anti-virus software on it (another way to have random performance hits). Its strictly used just for running the Shark. I run the power through a high-end surge protector, not as good at cleaning power as a UPS, but should be ok. The random nature of this does make it tough. And it only does it on very long jobs. What's odd to me is that the control panel appears to continue to run as when I find this 'paused' state, the screen shows that the job completed. The system does not require a reboot, power shut off etc, to run the next job. My thoughts tending towards heat, but I don't know how to test that. I'm hoping Tim will weigh in on this.

Thank you for thinking about this and offering ideas.

[tonydude]

Make sure you contact nextwaveautomation about this as they don't always read the forum. Call them up at support.

Tony

[tonydude]

You can either email at support@nextwaveautomation.com or give a call (419)491-4520. Available Monday-Friday 8:30-5:00pm Eastern Time.

Tony

[rungemach]

From the units I have seen, the Shark controller does not have any active air circulation (fans) and relies on heat transferring out through the case and some air exchange through a few small vent slots.

As there are heat generating electronics inside, the interior temp is bound to rise. Short jobs may not cause it to get as hot as it can over long continuous runs.
You may want to try and get some air circulation around and into the case, and see if your issues go away.

Excess heat is generally considered the enemy of electronics, unless they were made for those conditions.
As certain components get too hot they can become intermittent or just stop working.

[scubanimal]

Thank you all for the comments. All good ideas. I've captured a summary and sent it to support to see if they may have an idea. I'm glad to not hear this is a common issue, so I have hope this is solvable. I keep my controller on the top of the Shark Cage to give it the best cooling options, but have been wondering about the heat build up as I don't see any air vents on the bottom just a few on the back of the box, so I've wondered about air flow for some time.

Thanks again.

[howie0]

Have you had any response from the company on this? I am curious because I have had 10 and 11 hour jobs run fine but recently had a job just stop about 17 hours into a 27 hour job. Only mine wasn't like a pause, it was like it hit the end of the program, because the controller was on the first line of the program.

[rungemach]

The case for the controllers that I have seen only have a few small vents on one side. That does not allow much (if any) air circulation as it is difficult to get any air exchange using only one vent.
Another vent in an opposite side would help, a small fan would be even better. Either of those things may void the warranty. A fan blowing across the case may help, and would be better if you could direct part of that air flow into the vents

When I was a kid, there was a toy call the "easy bake oven" that basically used a small light bulb in a non vented case to get hot enough to bake small cakes with.

I do have a picture of the interior of a shark controller, and you can see the fanless power supply, the solid state relay, and the L shaped aluminum heat sink for the circuitry.
all of these item generate heat. They will heat the air in the case and in turn the case will try and radiate that heat out to the ambient air.
Some air circulating through the case would definitely aid in getting the heat out. Since 24 volt dc fans are available and cheap, I don't know why one was not built into the case.

If your case is notably warm to the touch after running a longer job, you may want to consider adding something to help its cooling.

[scubanimal]

Conclusion...sort of. I did contact the next wave support, and was told to reseat / replace the memory card. Looking at that I'm not sure how I'm going to get it out, so I've not tried yet. Mostly because it has not stopped since. I do now, on very long jobs, blow some compressed air in the vents of the controller, not hard but enough to exchange the air in it for a minute or so. I'll do that maybe twice during the day. I have noticed that if I've not done that for a while (like after a bunch of short jobs) there is a fair bit of dust. And I'm sure that dust will heat things up even more.