"0 mm of filament used in this print" [FIXED]

Hi all,

I've been attempting to learn how to use Skeinforge, and mostly failing. I've produced a few successful prints using Slic3r to generate my gcode, but I thought it would be good to get Skeinforge working as well. I think I may have a little bit figured out what the various settings are, but the issue is that whenever I slice an object, it invariably tells me that it uses 0mm of filament. I am not sure what would cause it to think this, as it seems to be planning proper extrusions, but I don't know for sure (sample file attached). Whenever I've opened Slic3r gcode in Pronterface it's given me an actual non-zero estimate, so I don't think its a Pronterface issue. Additionally, it's not generating the code for proper temps, similar to [forums.reprap.org].

For reference, I've tried two different version of SFACT, 45 and 50. Issue is the same either way. Don't know if it matters, but the printer is a Prusa i2, run by RAMPS with Marlin, and my computer is Ubuntu 12.04 64-bit.

Thanks!

Edited 1 time(s). Last edit at 12/19/2012 12:16PM by TopherMan.

Who is telling you 0 mm of filament - pronterface? There is actually a bug in the new gcoder.py module - it does not add in the filament that is consumed from the final E axis reset until the end of the print. I think for most average sized prints, skeinforge never resets the E axis, so this will result in 0 mm being reported. Slic3r seems to throw some axis resets in there, so it is only the final bit of the print that is not being accounted for.

I solved this by adding the following line of code to gcoder.py just prior to the return statement in routine filament_length:

total_e += cur_e

Thanks jbernardis... for bringing this issue/bug to the forefront and offering a fix for this. When I upgraded to Skeinforge v50 from v41, yes the Filament Used Qty. was magically no longer being reported to Pronterface. Hmmm...

I would like to fix Skeinforge to report filament usage to Pronterface as you have, however searching all Skeinforge folders and my boot drive (and hidden files) the file gcoder.py does exist.

Can you give us some more details about how to fix this ?

A very appreciative Skeinforge user.

Actually, the gcoder.py is within Pronterface. For me, it was at line 204.

Thanks, jbernardis!

Thanks TopherMan... for the information on where you found the gcoder.py file.

However after doing another extensive search, lots of gcode*_*.py files found, but NO gcoder.py file.
This is the correct file name (gcoder.py), right ?

ALL Printrun/Pronterface and Skeinforge folders, and my boot drive (with hidden and system files being searched also), the file gcoder.py apparently does exist on my PC .

Anybody have another suggestion as to why I cannot find this file, or why it may be missing from my Skeinforge v50 install, and how to locate and fix this ?

Sure would be nice to have Skeinforge report to Pronterface filament usage.

Thanks in advance for everyone's help.

Hmm, well, it's right there in the main Printrun folder, right next to pronterface.py. Are you using the newest version of Printrun? I looked around on Github and it looks like gcoder.py is a newer addition, and the calculations originally happened in pronsole.py. You could update to the most recent Printrun and edit gcoder, or you could see if you can edit pronsole to work.

Here's the diff where gcoder.py was added: [github.com]

Yep... your right TopherMan

File gcoder.py should be there in the printrun directory. Problem is apparently with the setup file(s) of Pronterface (with some setup/install downloads out there).

Found this very recent post on GitHub...
[github.com]

"Added missing files to setup.py"
"...install Printrun via setup.py, I don't have gcoder.py..."
[github.com]

Would you happen to have a solution (or a download link for setup/install with the gcoder.py file included) ?

Thanks !

Edited 1 time(s). Last edit at 12/19/2012 01:55PM by ShawnT98027.

Not sure... I'm on Linux, so all I have to do is clone the repo and run it. Maybe you could try the setup.py attached to the pull request you linked? Sorry I'm not more help, I've never bothered trying to set up any printer software on my Windows machine, so I don't actually know how to get python scripts running on Windows.

gcoder.py is a relatively new addition to pronterface, but I'm not sure the problem didn't exist in the old code too. I adapted this code for my own use, and I seem to remember having to make a change. Recently I changed my code to use the (fixed) gcoder.py instead of the old code, so I've lost track as to whether or not I had to change this previously.

In any event, you should probably download the latest printrun. I'm actually going to go to its github right now and report this as an issue.

Oh - and I'm glad to have been of help.

I´m a recent reprap user. Using Repetier-host V0.83 and Skeinforge to generate gcode for my prusa-mendel (Ecksbot).
I make my desings in Solidworks. It gives me the volume and mass of the 100% fill part. Even tough skeinforge uses the fill parameter (ex. 0.15), when I ask for Generate Statistics, the mass and volume are always wrong, always more than the real volume of the part. This way I can´t know the filament usage. I believe there is a bug, or maybe some wrong parameter in by settings. Do you guys get mass and volume right?

Other thing. I link the way Slic3r comments in gcode some parameters, and in the end the filament usage. Isn´t it possible to skeinforge to put this information commented in gcode?

Any help is appreciated.
Paulo

I think this thread was accidentally/way to prematurely marked [FIXED]... unfortunately I have not seen or found it to be fixed at all.

"Thanks jbernardis... for bringing this issue/bug to the forefront and offering a fix for this. When I upgraded to Skeinforge v50 from v41, yes the Filament Used Qty. was magically no longer being reported to Pronterface. Hmmm...

I would like to fix Skeinforge to report filament usage to Pronterface as you have, however searching all Skeinforge folders and my boot drive (and hidden files) the file gcoder.py does exist.

Can you give us some more details about how to fix this ?

A very appreciative Skeinforge user."

Not sure how to do this, but please remove the [FIXED] status of this thread, so hopefully the problem really can be addressed and actually fixed. And the fix is posted.

Thanks in advance for who ever does fix this issue and POST the FIX.

Shawn,

For some reason I can't seem to edit my posts, so I can't change it at the moment. However, I am planning to change it to WORKAROUND, not remove it completely. The issue does not lie within Skeinforge, the issue is Printrun. See [github.com]. The solution is to get the newest version of Printrun and edit gcoder.py, so in my mind it's solved, though it does require some hacking. Hopefully the Printrun folks will fix that soon. It looks like the issues stemming from setup.py are eliminated (at least, issue 300 is listed as closed), so pull the latest version, add the line to gcoder.py, and you should be set to go.

Paulo, I tend not to look at the mass and volume calculations. All of these measurements have certain underlying assumptions that can vary from setup to setup (back pressure in nozzle, extrusion shape, etc), so I suspect they are not terribly accurate. Additionally, I don't think the fill parameter takes perimeters into account; aka, an object with 10% fill will actually be something more like 20% full when you include the perimeters.