Firmware Error. Print Head Freezing during print

[taikonaught 2]

Hi all,

 

I have an Ultimaker 2 that is freezing very predictably at the exact same spot every print.

I have been trying to print a small part, and nearly half way through layer 2 the print head just stops. The the fans stays on, the hot end cools down. It is also responsive to homing commands.

 

It tells me:

 

Your printer's firmware reported an error. Due to that the ongoing print job will be cancelled. Reported error: Printer stopped due to errors. Fix the error and use M999 to restart. (Temperature is reset. Set it after restarting)

Send: N554 G0 F5400 X127.866 Y101.235*67
Recv: ok
Send: N555 G1 F1800 X101.235 Y127.866 E82.37168*49
Recv: Error:Printer stopped due to errors. Fix the error and use M999 to restart. (Temperature is reset. Set it after restarting)
Changing monitoring state from "Printing" to "Cancelling"
Send: N556 M108*44
Recv: Printer stopped due to errors. Fix the error and use M999 to restart. (Temperature is reset. Set it after restarting)
Recv: ok

Just for a bit of background, the LCD screen on the machine has not worked for years, so I have been using Octoprint, which has been fine.

I have tried printing directly from Cura via USB and exactly the same thing happens. I have tried printing other parts from Gcode that have successfully printed before, and the same thing happens.

 

The machine was working fine with an older version of the firmware installed. After this problem happened, I then went to install the latest firmware, but the problem persisted.

 

Any help would be greatly appreciated!

Thanks.

 

 

[taikonaught 2]

i have tried different ways of printing from different gcode.

Printing direct from Cura over USB, tried slicing it in new and old versions of Cura, tried printing via Octoprint and Pronterface, all wit the same result.

 

Any clues? Anyone?

[Torgeir 280]

Hi @taikonaught,

 

I'm sorry to see that you've had this problem for so long.

 

I wonder why is there no "valid" error code as we see it on the LCD display to be seen on Octoprint?

 

An UM2 have just a few errors that will stop the printer with a given message:

 

We have temp probe error, heater error and the three stop switches (X,Y and Z), if my memory serve me right..

 

In your case I'll think it's either a temp sensor/heater head error, or a temp sensor/heat bed error.

 

If the temperature drop too much offset during print the printer report temp error, also if the temp sensor have an intermittent connection the same is reported and the printing stop due to this error.

In the LCD display you're asked to go to an Ultimaker web address for guide.

 

I'm not sure if any of this is reported to Octoprint.

 

You might also try to use blue tape on bed without using heat on the bed, this in order to eliminate the heat bed problem.

Although, a direct probe intermittent wiring break might still stop the printer.

 

You may also try to let the printer print without filament (use say 180 deg. C for head), start printing without filament and bed a little offset down. In this situation, the head should easily keep the temperature steady.

Also, this will show if this error have something to do with the height..

 

Just something to try.

 

I've also been looking a little to the OLED wiring, but we may wait a little with this.

 

Good luck.

Thanks

Torgeir

[taikonaught 2]

Hi Torgeir,

 

thanks heaps for the ideas. Unfortunately, I tried them all without success.

Octoprint has good temperature monitoring. See image below.

I tried printing with no heated bed, and without filament, but it froze at the same spot every time.

 

The fact that it freezes at exactly the same spot to the mm makes me think that it isn't a sensor issue. You would expect it to stop at a slightly different point in the print.

 

I did get another error (M141 S28).

 

Is there any chance someone could slice the attached STLs and send me the gcode? Just in case I have a bad setting somewhere in Cura (even though it is failing prints from Gcode is has printed successfully before).

 

Is there a way to do a factory reset on the printer?

 

Thanks

 

 

 

 

On 6/30/2021 at 4:25 AM, Torgeir said:

e had this problem for so long.

 

 

base 1.STL base 2.STL

[Torgeir 280]

Hi @taikonaught,

 

Aha, -think the "Heated Build Volume" is enabled for UM2 in Cura.

 

Look in the printer setting and verify this setting is not enabled.

 

Just like this:

 

 

If we're lucky, this may work..

 

Thanks

Torgeir

 

[Torgeir 280]

Hi @taikonaught,

 

Forgot to slice the files, but here they are:

 

Done with Cura 4.9.1 (fix): Selected res. "Fine 0.1" with Added changes; Skirt and combing mode set to "not in skin".

 

UM2_955473258_base1_&_2.gcode

 

UM2_955473258_base1_&_2.3mf

 

Thanks

Torgeir

 

[taikonaught 2]

Thanks for the idea, but that setting was not enabled.

That error did pop up, but it isn't the error that froze the print.

 

Interestingly, I was able to print one file from old gcode. Just a little cube to test print temperatures on.

This makes it even more confusing...

 

Any other ideas?

 

On 6/30/2021 at 4:25 AM, Torgeir said:

e had this problem for so long.

 

 

[taikonaught 2]

2 hours ago, Torgeir said:

Forgot to slice the files, but here they are:

Thanks for doing that, I appreciate it.

Unfortunately, it froze at precisely the same place of layer 2 as the others did.

[taikonaught 2]

This one came up:

SerialException: write failed: [Errno 5] Input/output error

Does that shed any light on the problem?

 

Thanks and sorry!

On 6/30/2021 at 4:25 AM, Torgeir said:

e had this problem for so long.

 

 

[Torgeir 280]

Hi @taikonaught,

 

This was not that easy..

 

So I went back to your first mail, -in here there is a gcode that does not "exist"..  M108 -not used in Marlin (UM2).

 

So I've dig a little around and found that this command is something that's added by Octoprint(?).

 

Have a look at this from "Github -/gcode/issues":

 

https://github.com/KevinOConnor/klipper/issues/1009

 

If deselecting this gcode "cmd" in Octoprint help, no idea, but maybe worth a try?

 

Torgeir

 

[Torgeir 280]

11 hours ago, taikonaught said:

This one came up:

SerialException: write failed: [Errno 5] Input/output error

Does that shed any light on the problem?

 

Thanks and sorry!

 

 

Well, -when you renewed the firmware, one of the changes was increasing the baud rate up to 250000 from before 115000.

Under this statement there was a note:

 

"The default baudrate is 250000. This baudrate has less jitter and hence errors than the usual 115200 baud, but is less supported by drivers and host-environments."

 

Your problem started before firmware upgrade, right.

 

---

It looks like this error can trigger different "false" errors, so not an easy one..

 

If the interrupt level for this error is set to too high priority, it might stop some processing that's make the printer go to "time out" and stopping the printing process..

 

Something strange is happening for sure..

 

Thanks

Torgeir

 

Edited July 2, 2021 by Torgeir
Text error again...

[Torgeir 280]

Hi @taikonaught,

 

Do you print from the SD card on the printer?

As you've managed to print a small cube without this fail, -what media did you print from?

Was this an old file (the small cube) on the same SD card?

By some reason the printing process just jump back to an invalid line -and this is a fatal error that stopped the printing.

 

Just a shot..

Torgeir

 

[taikonaught 2]

7 hours ago, Torgeir said:

Hi @taikonaught,

 

Do you print from the SD card on the printer?

As you've managed to print a small cube without this fail, -what media did you print from?

Was this an old file (the small cube) on the same SD card?

By some reason the printing process just jump back to an invalid line -and this is a fatal error that stopped the printing.

 

Just a shot..

Torgeir

 

Hi Torgeir,

 

the cube was an old, previously sliced file stored on the D drive of my computer.

But other files in the same location (that have printed successfully in the past), that were sliced at the same time, from the same version of Cura all seem to fail.

 

I don't know what is so special about this cube. The block is only 20 x 20 x 10mm. I used it to test printing temperature and speed. I bought some filament that didn't seem to be extruding properly, so I increased the temperature using this as a test piece. I had changed the Cura flavour to Rep-Rap, as the normal Ultimaker flavour did not let me directly set the temperature. I thought that might have been the issue, but I have since changed it back to Ultimaker, and even printed the files that you sliced for me, and the problem is still there.

 

The fact that the prints always fail at exactly the same spot really makes me wonder.

 

 

 

 

[taikonaught 2]

Interestingly, it seems to always print layer 1 successfully, then fail on layer 2.

I found one other sliced file that prints ok as well. This one was sliced just a couple of weeks ago, and is more complicated than the cube (5 knurled, threaded nuts.) No idea why that one is ok, while the others are not.

 

Is there a process that happens at layer 2? like the board pinging the hot end or something? Even quite complex 1st layers seem to work fine

 

This is another one that printed ok before , but now fails at the beginning of layer 2.

 

 

8 hours ago, Torgeir said:

Hi @taikonaught,

 

Do you print from the SD card on the printer?

As you've managed to print a small cube without this fail, -what media did you print from?

Was this an old file (the small cube) on the same SD card?

By some reason the printing process just jump back to an invalid line -and this is a fatal error that stopped the printing.

 

Just a shot..

Torgeir

 

Hi Torgeir,

 

the cube was an old, previously sliced file stored on the D drive of my computer.

But other files in the same location (that have printed successfully in the past), that were sliced at the same time, from the same version of Cura all seem to fail.

 

I don't know what is so special about this cube. The block is only 20 x 20 x 10mm. I used it to test printing temperature and speed. I bought some filament that didn't seem to be extruding properly, so I increased the temperature using this as a test piece. I had changed the Cura flavour to Rep-Rap, as the normal Ultimaker flavour did not let me directly set the temperature. I thought that might have been the issue, but I have since changed it back to Ultimaker, and even printed the files that you sliced for me, and the problem is still there.

 

The fact that the prints always fail at exactly the same spot really makes me wonder.

 

 

 

 

[taikonaught 2]

Oh, and printing from the SD card has the same problem...

[Torgeir 280]

Hi @taikonaught,

 

Hmm, well you might try to restore to default factory setting, as you suggested in here -may by you've already done it?

If not, you just send the cmd: M502  -then save the factory setting by using cmd: M500.

 

It should also be possible starting a SD card logging file.

As done here; from the implemented G code used in UM2.1 firmware (https://github.com/Ultimaker/UM2.1-Firmware)

 

M928 - Start SD logging, i.e. all commands output is written to the specified file (M928 filename.g) - ended by M29

 

I've never done this before, but should work.

Let's see if this give us something..

Torgeir

 

 

 

 

 

 

[taikonaught 2]

Hi Torgeir,

 

So I did a factory reset, and now the the bed drives up into the print nozzle at speed. I'm lucky it didn't break the glass...

 

I was able to protect the glass until the print got to the point that it was failing at, and it failed just as before 😞

 

So now my problems are two fold. It feels like the machine now thinks it is an Ultimaker 2 Extended. I want to cry.

 

 

[Torgeir 280]

Hi @taikonaught,

 

This mean that your last firmware was made for an UM2E.

This is an easy fix to do, as the only different between thous two is the Z height.

A standard UM2 use Z (number of full height) = 200 (and the E version have Z=400).

 

To change just this parameter, do as follows:

Send cmd: M92 Z200.00    (This will change the height to 200 full step)

Then send cmd: M500      (This save this setting into EEprom)

 

Using M502 will put back all factory default setting to EEpeom.

 

Using M501 will read EEprom setting for you to see.

Using M503 will read current memory setting for you to see.

 

Using M502 will always bring back the last "updated firmware setting"-

 

 

For a permanent good factory default setting, you'll need to upgrade firmware again with the correct UM2 version printer you have.

 

So a few important questions, are your printer upgraded or is it a plain standard UM2?

 

A standard UM2 have a black feeder unit on the back and you cannot change the nozzle size, as it is a fixed 0.4 mm nozzle.

Right?

 

The easiest way of fixing this for you is to download a "new" firmware for your printer from Cura, but be sure to select the correct version of the UM2 you have -and all should be good!

 

If the original problem still persist, we'll make a log file to have a better idea of what's going on.

 

Thanks

Torgeir

Edited July 6, 2021 by Torgeir
More info.

[RealShe_Cheer 0]

On 7/6/2021 at 4:00 AM, Torgeir said:

Using M501 will read EEprom setting for you to see.

Hi Torgeir,

How can I view this information in Cura 5.6.0?

Edited January 1 by RealShe_Cheer

[Torgeir 280]

Hi @RealShe_Cheer,

 

Welcome in here and Happy New Year.

 

Well, We'll need to know what printer you have as some printers may not have this commands implemented.

Cura does not have an easy way to do this.

Lots of people use OctoPrint, here: https://octoprint.org/blog/2022/05/17/new-release-1.8.0/

Or Arduino, but this latter one is a little harder to use. I'll think OctoPrint are the best one for this issue..

 

Here is a tread covering a few different approaches to send a command and have a feedback from the printer.

https://3dprinterly.com/how-to-send-g-code-to-your-3d-printer-the-right-way/

 

In here are lot's of people with "all" kind of 3D printers, so telling what printer you have and your problem with it may help a lot.

 

Thanks

Torgeir