UM2(+) known bugs?

[CarloK 190]

Soon a new release for the UM2+ will be created to add the Tough PLA material. In this topic I want to investigate the known problems that we can perhaps fix and add to this new release.

 

At GitHub there is for the UM2+ only 1 reported issue:

- A problem with filament grinding after material change.

 

The UM2 repository at GitHub has a few more issues, I will look into those.

 

I'm surprised no other issues have been logged. For example, when testing the material import function I found several bugs (these are fixed now and will be in the next release).

 

Do you people know of more bugs? Then please list them here or on GitHub.

No promises that we can squeeze them in the next release, but we will try. Please, only bug reports. Since the UM2 is at end-of-development stage we will only address bugs and issues that enhance the print quality.

[CarloK 190]

4 weeks have passed and no replies.... Everybody is happy with the UM2(+) as it is?

 

Here are a few things that will be in the oncoming release:

Copied stepper module fixes from UM3 / Open source Marlin project:

These fixes solve the problem where during fast movements you sometimes heard a loud tick sound like some mechanical hiccup was occurring.

- Fix: the multiply routine could have an overflow, solved by adding a 24x32-bit routine
- Fix: ensure the stepper interrupt triggers immediately again when it takes too long.

 

Improve material import:

- Fix: a material name length of 8 would crash the printer. Now consistently allow a length of 8 characters everywhere.

- Fix: illegal values for the material retraction-length and retraction-speed are now set to default values.

- Change: Improve error handling for very large positive or negative numbers.

 

Other changes:

- New: added Tough PLA (TPLA) material profile.

- Fix: retract at end of print. Contributed by TinkerGnome.

- Fix: another bug with retraction at end of print. If we were already retracted (G10), then we were retracting too much.

- Fix: M117 was missing a character when printing via USB. Cherry-picked from UMO repository. Contributed by AmedeeBulle

- Change: Changing material during a print is made 15 to 30 seconds faster by removing an unneeded filament heat-up (the filament is already hot).

- Change: several menu's where the only option was to 'Click for return' now have the focus set to this button. This removes the need to first rotate the button and results in faster user experience.
 

PLEASE REPORT PROBLEMS!!!!

In the forums I see people complaining about no new firmware releases coming out, but when no problems are reported then we can't fix them.

Report your problem at github or post it here, then I'll add it to github.

Edited August 13, 2018 by CarloK

[KiwiMrDee 0]

after aborting a print (UM2) the filament retracts meaning unless i manually progress it, the next print will start printing air until the filament reaches the hot end 

 

[CarloK 190]

@KiwiMrDee Thanks for your report. This is a known problem and has partially been fixed by the two above mentioned retraction fixes. The only known problem situation is where you first pause and then abort a print.

 

We will investigate this further, but I have limited time to look into the UM2 problems. The Cura 3.4 beta release is nearing, so perhaps we release the current list of improvements and move further improvements to a future release.

[Link 32]

I was going to mention this 

 

At GitHub there is for the UM2+ only 1 reported issue:

- A problem with filament grinding after material change.

 

has this one been fixed as it’s not listed in the list of fixes in the next release ?

 

thanks 

[Link 32]

Also, could we get some more information on the screen whilst a print is running, like the tinker firmware. I am always nervous about installing 3rd party FW so an official UM version with nozzle temp, build plate temp etc would be great !  

 

[Nicolinux 288]

I can't remember correctly how this bug "worked" but there was something about the print head freezing in place at the end of a print. It was about parsing .gcode files and the fix was to add enough random characters to not trigger the bug. Does it ring a bell?

[Link 32]

No, that was a different issue, that was blobs at the end of the print, when the print finished the extruder just kept pushing material due to the length of the gcode at the end. 

 

This is different and is when you change material, rhe extruder pushes filament before it retracts and that initial push is so hard it grinds filament. 

[CarloK 190]

26 minutes ago, Nicolinux said:

I can't remember correctly how this bug "worked" but there was something about the print head freezing in place at the end of a print. It was about parsing .gcode files and the fix was to add enough random characters to not trigger the bug. Does it ring a bell?

I guess you mean the bug reported here. That bug created a blob at the end of a print and only happened with Cura v3.2.

This wasn't mentioned in my list of fixes but is there as well and was fixed in Cura 3.3 too.

 

There is also another fix for the quickstop function where the position of the e-axis was set wrong. This is a 3rd fix to the filament position at the end of a print. Submitted by Tinkergnome.

[Link 32]

10 minutes ago, CarloK said:

I guess you mean the bug reported here. That bug created a blob at the end of a print and only happened with Cura v3.2.

This wasn't mentioned in my list of fixes but is there as well and was fixed in Cura 3.3 too.

 

There is also another fix for the quickstop function where the position of the e-axis was set wrong. This is a 3rd fix to the filament position at the end of a print. Submitted by Tinkergnome.

 

What about the one mentioned relating to the material change ?, is this one fixed ?

 

At GitHub there is for the UM2+ only 1 reported issue:

- A problem with filament grinding after material change.

[CarloK 190]

5 hours ago, Link said:

At GitHub there is for the UM2+ only 1 reported issue:

- A problem with filament grinding after material change.

 

has this one been fixed as it’s not listed in the list of fixes in the next release ?

The specific situation as reported in that issue has been solved by the 3 already mentioned fixes to filament position at the end of a print. However, I haven't closed the ticket yet since I found another problem: when you pause a print and then abort, the filament is retracted too far. 

 

This problem is giving me headaches... I've looked at it for a whole day and then had to work on other projects again. It still has my attention but I don't get a lot of time to support the UM2. Right now I'm thinking about creating a release with all above fixes so you people at least get several improvements. And, ... it is open source. Everybody can look into the program and contribute improvements...

[Link 32]

3 minutes ago, CarloK said:

The specific situation as reported in that issue has been solved by the 3 already mentioned fixes to filament position at the end of a print. However, I haven't closed the ticket yet since I found another problem: when you pause a print and then abort, the filament is retracted too far. 

 

This problem is giving me headaches... I've looked at it for a whole day and then had to work on other projects again. It still has my attention but I don't get a lot of time to support the UM2. Right now I'm thinking about creating a release with all above fixes so you people at least get several improvements. And, ... it is open source. Everybody can look into the program and contribute improvements...

 

thanks for the update, appreciate you guys have lots to do.

 

Will the firmware update be made with the next release of Cura or can you/do you release firmware on its own ?

 

thanks

[CarloK 190]

5 hours ago, Link said:

Also, could we get some more information on the screen whilst a print is running, like the tinker firmware. I am always nervous about installing 3rd party FW so an official UM version with nozzle temp, build plate temp etc would be great !  

 

That's a new feature request and was rejected many times before since it conflicts with the Ultimaker strategy where the printer should be as easy as possible to use.

My personal opinion is that Ultimaker should also cater to the tinkerers. One way to do that would be by adding a menu setting where you can enable an 'advanced menu' feature. This will be considerable effort and since the UM2 is an older product we won't get time to create that. Tinkergnome's software has proven itself to be a good alternative for the tweakers. It is almost impossible to brick your UM2. Try the alternative and you can always revert to the original firmware.

[CarloK 190]

4 minutes ago, Link said:

Will the firmware update be made with the next release of Cura or can you/do you release firmware on its own ?

My intention is to have the current state included with the upcoming Cura 3.5 beta release. Otherwise we'll post a download link here on the forum.

[Link 32]

Just now, CarloK said:

My intention is to have the current state included with the upcoming Cura 3.5 beta release. Otherwise we'll post a download link here on the forum.

 

many thanks, appreciate your time

 

I was worried about the tinker firmware missing some of the parts from the official firmware and somehow affecting performance of the print or worse still causing a crash etc. I can't quantify how the tinker changes are made, maybe its not possible for it to affect print quality as its only in the UI side ?

[MGSR 0]

Hi there,

I encountered a problem with printing large files. Whenever I attempted to print g-code files larger than 500 MB, the printer (UM2+) does not complete the job. He doesn't throw any error message or warning, he just stops printing at some point. And this point looks like it would correspond to the data volume of about 500 MB (I print a 600 MB file and get a model completed to about 5/6). I tried this and could reproduce this "error" two times, I am not trying any further since it wastes a lot of filament. The G-code itself looks fine, it appears the printer is just not reading all of it...
A cannot share the g-code as it proprietary and scientifically sensible material.
My workaround so far is just to simplify the model until the file size is lower than 500 MB, which is a pity since i lose details.
If printing such large files is not possible, the firmware should at least warn the user...
Cheers
Michael

[CarloK 190]

@MGSR printing files of 500Mb in size shouldn't be a problem, we tested the printer with files op to 1.3Gb without problems. You are the first person to report this problem so it is not a common problem. Something in your environment or way or configuration causes this problem.

 

Finding the cause is difficult when you can't share the gcode.

 

My first suspect is a problem with your SD card. It is very easy for the file system to get corrupted, for example one time ejecting the card without first selecting the 'eject' function on your computer is enough to cause these problems. It might also be an SD card brand that's more susceptible to noise or other data corruptions.

 

Could you re-try with a well known SD card brand? And first fresh format the card with then only this single file on the card?

[Torgeir 212]

Hi Carlok,

 

Thanks for your reply, I've just made a "new" tread..

And yes, firmware is the other side of the chain as important as the slicer..  :)

I'd wish there was a hidden stepper menu, that could be altered for the UM2 users, this hardware is -not very far apart..

 

Thanks allot for your work.

 

Best regards

 

Torgeir 

[CarloK 190]

15 hours ago, Torgeir said:

I'd wish there was a hidden stepper menu, that could be altered for the UM2 users, this hardware is -not very far apart..

Could you elaborate a bit more, since now I think I understand you but am not sure. Do you mean the UM2 and UM2+ are very similar? And then, what settings would you like to see in that new stepper menu?

[Torgeir 212]

Hi @CarloK

 

Well, if I'm right; the UM2+ have E369.00 (using a 200 step/rev stepper motor).

 

UM2 have E282 (using a 400 step/rev stepper motor)

 

As an example:

UM2+ E369.00

UM2   E282  

INV (a check box to change direction for UM2 if needed)

If UM2+ is selected the direction will go to default for  UM2+

 

The upgrade kit for UM2 to become UM2+

 

Software related issues:

1)  New stepper motor 200 step/rev and changed direction and E step.

2)  New heath block, Ohlson block type, slightly different routine "PID". 

 

This is all software changes as far as I know.. I'm using the Ohlson block, with same PT100 sensor and same heater 25 W 24 VDC as original. I have not noticed any difference with the kind of objects I'm printing. 

Also thing worth notice is that, on the later Cura versions (if you select UM2), Cura ask if you have installed Ohlson block. So this should be good I'll think.

 

The other parts of this modification is hardware as far as I can see.

 

Thank you for keeping our firmware in good shape.  :)

 

Regards

 

Torgeir

 

[Nicolinux 288]

Here is another one - spotted it on the German sub-forum. This seem to happen when the file material.txt is malformed (check out the video):

 

[CarloK 190]

Yep, could very well be a material.txt related problem. Most likely the name of the new added material was 8 characters or longer and then weird things will happen. Solution is to reset to factory settings.

[Link 32]

On 8/23/2018 at 9:32 PM, CarloK said:

The specific situation as reported in that issue has been solved by the 3 already mentioned fixes to filament position at the end of a print. However, I haven't closed the ticket yet since I found another problem: when you pause a print and then abort, the filament is retracted too far. 

 

This problem is giving me headaches... I've looked at it for a whole day and then had to work on other projects again. It still has my attention but I don't get a lot of time to support the UM2. Right now I'm thinking about creating a release with all above fixes so you people at least get several improvements. And, ... it is open source. Everybody can look into the program and contribute improvements...

 

Hi @CarloK any news when a release might be coming out with these firmware fixes ?

 

Many thanks

[ahoeben 1,767]

Based on activity on github, I can predict there will be a beta-release of Cura 3.5 fairly soon, probably even as soon as coming week.

[CarloK 190]

The new firmware will be included in the Cura 3.5 Beta. Since that's not my department I don't know the release data details but they are ironing out the last issues.