Unretraction issues



  • I'm on my 2nd conversion of a Tevo Little Monster and having an issue with retraction; well, actually unretracting. The only difference in components from my first is using an E3D 13N-CM instead of a no-name pancake off Amazon. That in mind, after calibrating the extrusion lengths I doubled my M92 E value (which makes sense since the E3D are .9 whereas the others are 1.8 steps):

    Tevo 1: M92 X160 Y160 Z160 E860
    Tevo 2: M92 X160 Y160 Z160 E1720

    The retraction is fine from what I can tell (not sure how to really measure it, I just see the hob gear turning in reverse for a short stint and no stringing), but when it hops to the next line it doesn't really unretract and starts to print … nothing ... for a few mm until it's primed again. This is really bizarre. I'm not really sure what to do. For the time being, I've put the stock Tevo stepper back on (E value of 880 btw), no change to the retraction settings, and it's printing fine. I thought it might be a faulty stepper that pukes on load so I tried another (I picked up 3 for the 3 printers I'm putting together) and still same results. Is there some black euro magic I need to perform to get these to work right?! I'm tempted to just purchase 3 more Amazon pancakes.


  • administrators

    Have you tried reducing extruder acceleration?

    Do you have the specifications (in particular, rated current and inductance) for the new motor?

    Are you using 12V or 24v power?



  • It's 24V and the only thing I changed was the M92 line in the config.g when calibrating the extrusion length before operating. I did try increasing the amperage from 1 (the Amazon pancake is rated at one, so I lowered from the default of 1.8) to 1.4 since it was a bit bigger of a stepper. It didn't make any difference, so I set it back to 1. Before I started digging any further and making a mess of my configurations, I hit the forums.


  • administrators

    @bLiTzJoN:

    It's 24V and the only thing I changed was the M92 line in the config.g when calibrating the extrusion length before operating. I did try increasing the amperage from 1 (the Amazon pancake is rated at one, so I lowered from the default of 1.8) to 1.4 since it was a bit bigger of a stepper. It didn't make any difference, so I set it back to 1. Before I started digging any further and making a mess of my configurations, I hit the forums.

    As per my previous question: do you have the specifications of the E3D motor that you replaced the Amazon motor with?



  • I'm kind of having the same issues. What are some good baseline settings for accel/jerk for an extruder motor? I've never had to configure these for the extruder. Here's what I have

    M201 X2500 Y2500 Z100 U100 E1000 ; Accelerations (mm/s^2)
    M203 X180000 Y180000 Z1800 U1800 E3600 ; Maximum speeds (mm/min)
    M566 X800 Y800 Z30 U30 E20 ; Maximum jerk speeds mm/minute

    Unfortunately, I don't have any specs on my steppers. I bought them a while ago and there are no labels on them

    Edit: Found them:
    1.8 stepper
    0.8A Current
    Inductance per phase 7.5mH 20%



  • Here's info on the E3D stepper. To clarify, I'm comparing 2 printers. I have one Tevo Little Monster equipped with an Amazon stepper and another with the E3D stepper. The biggest difference between the two is the steps, the E3D is double the resolution which is why I had to double the value for M92. I need to confirm this, but the behavior feels like if I have my retract set to 4mm it will retract 4mm but only unretract 2mm when resuming. Is there another config value I have to double due to the resolution? The original stepper is 1.8, the Amazon stepper is 1.8, and they work fine.



  • Too much microstepping for .9 motors? If the original 1.8 motor worked that could be the case.

    Maybe the feedrate for unretract is too fast?



  • Are you using firmware retraction? Also, can you type M207 into gcode console and post the output?



  • Here's mine:
    M207
    Retraction/un-retraction settings: length 2.00/2.00mm, speed 1000/1000mm/min, Z hop 0.00mm

    Does that mean it is retracting 2mm instead of the 5.5mm I set in my slicer?



  • @SpeedyDad:

    Here's mine:
    M207
    Retraction/un-retraction settings: length 2.00/2.00mm, speed 1000/1000mm/min, Z hop 0.00mm

    Does that mean it is retracting 2mm instead of the 5.5mm I set in my slicer?

    It depends on the slicer and if the slicer has firmware retraction enabled. If firmware retraction is enabled in the slicer, then your retraction settings are what is output by the M207 code (assuming the slicer enables g10 / g11 commands for retraction). If slicer does not have firmware retraction enabled, it will be using the settings from the slicer. I was curious to see if firmware retraction was being used and if someone put in a bad R value in the M207 command line.

    Rnnn positive or negative additional length to un-retract



  • I'm using S3D, so I'll have to research what is really happening. I don't believe it is using firmware retraction because I can adjust it in S3D and impact the output. If I lower the retraction to nothing or close to nothing but it does ok. I'm almost finished putting together my 3rd Tevo LM and will start with the E3D stepper that way I will have a printer with all 3 options (Amazon, Stock & E3D) to play with.



  • It's easy to check if the slicer is using firmware retraction. Just check the gcode file for G10 (with no other parameters) commands. Each one should also be followed by a G11.

    Edit IIRC S3D does something differently and uses some other command instead of G10/G11. Can't remember exactly what the work around involved and later versions may not exhibit the same behaviour.



  • Even easier in the process settings of s3d check the G-Code tab and see if allow M101,M102 and M103 option is ticked if so then you have firmware retraction on but AFAIk you have to set all the retraction option to as low as you can get for it to be effective.

    HTH

    Doug



  • I do not have that checked and, you're right, it does not use G10/G11.



  • Are you sure about that Dougal? I don't think M101, M102, and M103 relate to firmware retraction in S3D?



  • yes they do it is the old way of doing it there is a write up about it somewhere on the Reprap forums and it does work I use it all the time. Generaly it is deprecated but DC added support into RRF to allow them to be treated correctly.

    see http://reprap.org/wiki/Gcodes#M101:Turn_extruder_1_on.28Forward.29.2C_Undo_Retraction



  • Whatever steps/mm setting is, retract/unretract should be in sync: length might be incorrect, but it should be the same in both directions.

    Did you try at a lower speed? It may miss a few steps when doing unretract…

    Could you also try to slowly increase the retract/unretract value, and see if there is a point when it suddenly start to fail?

    Also, have a look at the G-Code fil to check if both values are the same! Don't you have an additional unretract negative value which may mess up things? If all seems ok, try to send both commands by hand, and measure how the filament moves after each command.


Locked
 

Looks like your connection to Duet3D was lost, please wait while we try to reconnect.