During my last print I noticed this jumpy behavior during heat up (blue is bed, red is hotend):

Graph of heatbed and hotend temperature over time

So I am wondering about the jumps. What could cause this? I do not think that it is a defunct sensor (it is in both sensors) and I also don't think it is real. Could this pose some sort of hazard as temperature might not properly be controlled?

I use an Anet A8 printer with Marlin 1.1.9 and Octoprint

  • 1
    $\begingroup$ Looking at the graph, it looks like OctoPrint is missing temperature readings from the sensor considering that a) thermal runaway protection is not kicking in, b) the heating curve continues the path it has set in by recovering to the position it should be after failed readings/measurements. You could try to change the USB cable between OctoPrint and printer board. $\endgroup$ – 0scar Apr 14 '19 at 9:45

Since the terminal temperatures never exceed the setpoints, there's no apparent potential for disaster. Whether it's due to missing readings or to some sequencing of power (current) applied to the bed vs. the hotend, it really doesn't matter.

If you have a similar graph of the temperatures over an hour of printing and you see signifcant anomalies there, that might be of interest.


It looks like everything is ok with the real temperature, and it is just Octoprint missing some readings, like Oscar said. But I think it is unlikely the USB cables fault, the whole print would fail in that case. I think the PI is to busy, maybe due to a Webcam streaming at high rate (try reducing the framerate). Another reason might be if you use a PI Zero W, especially if you transfer data over the Wifi at the same time, since Wifi and USB cause bottlenecks for each other on that board.

PS. I'd comment, but don't have privilege yet.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.