Jerky XY motion

Discussion in 'Calibration, Help, and Troubleshooting' started by Miasmictruth, Apr 10, 2016.

  1. Miasmictruth

    Miasmictruth Well-Known Member

    Joined:
    Sep 4, 2015
    Messages:
    804
    Likes Received:
    118
    Ok so I wrote out a couple gcode scripts the first is to show linear motion, I had meant to run it at 100mm/s on the first video but had the feed rate at 150% so I believe that should be 150mm/s the motion is nice and smooth



    Then I took the gcode for a circle with a diameter of about 41mm and stripped the Z out of it. I ran it 37.5mm/s for 3 circuits which was okish 66.6mm/s for 2 circuits which was not good then 100mm/s which is terrible

     
  2. Miasmictruth

    Miasmictruth Well-Known Member

    Joined:
    Sep 4, 2015
    Messages:
    804
    Likes Received:
    118
    The above videos were all run on RC3
     
  3. Alex9779

    Alex9779 Moderator
    Staff Member

    Joined:
    Sep 4, 2015
    Messages:
    2,411
    Likes Received:
    735
    Ok it would be helpful if you describe what you mean in word too. Videos and pictures have room for interpretation.
    But I think I know what you mean.

    First you do not think it is a slicer problem.

    Second it is not a problem limited to RC4 or RC5 but it is also in RC3 aka the official BB firmware.

    Third it is the strange side sounds you hear on that circle and I think I had that too in circles like I write in my other post...

    Is that correct?
     
  4. Miasmictruth

    Miasmictruth Well-Known Member

    Joined:
    Sep 4, 2015
    Messages:
    804
    Likes Received:
    118
    Yes that seems like a good summation.

    In words: As it travels around the circle it starts to hesitate/stop/jitter at certain points. These points most notably seems to be when one of the steppers has to reverse motion to which there are 4 such spots on a circle. When the hesitation happens it ends up destroying the outer perimeter making the print useless.

    It seems if I run the big box slow enough the probable diminishes or goes away which seems to be maybe around 30mm/s which to me seems to me like the motion system shouldn't choke at that speed, I should think I should be able to at least be able to slice at 50mm/s
     
  5. Miasmictruth

    Miasmictruth Well-Known Member

    Joined:
    Sep 4, 2015
    Messages:
    804
    Likes Received:
    118
    30mm/s failed... Trying 25mm/s
     
  6. Alex9779

    Alex9779 Moderator
    Staff Member

    Joined:
    Sep 4, 2015
    Messages:
    2,411
    Likes Received:
    735
    @Miasmictruth can you post your test files so I can test them too tomorrow?
     
  7. Miasmictruth

    Miasmictruth Well-Known Member

    Joined:
    Sep 4, 2015
    Messages:
    804
    Likes Received:
    118

    I am currently running your RC4 now, Here is a link to the object that kicked all this off.

    http://www.thingiverse.com/thing:1480191

    Here is a link to the current S3D settings I am using, beware I have a 0.8mm nozzel on right now as I wanted thicker walls for the container. Hopefully it works.

     
  8. Miasmictruth

    Miasmictruth Well-Known Member

    Joined:
    Sep 4, 2015
    Messages:
    804
    Likes Received:
    118
    No idea why that keeps posting as media and not a like to the file... Sorry
     
  9. Miasmictruth

    Miasmictruth Well-Known Member

    Joined:
    Sep 4, 2015
    Messages:
    804
    Likes Received:
    118
    Here is a couple prints in PLA the left was at 25mm/s the right was 20mm/s

    The side with the messed up extrusion is when y is reversing, the smooth side is when X is reversing.
     

    Attached Files:

  10. elmoret

    elmoret Administrator

    Joined:
    Mar 1, 2014
    Messages:
    631
    Likes Received:
    101
    That is a ballpark, they do not need to be that high in practice. Best to tune them.

    Currently I'm running 0.576/0.253/0.654/0.510v (XYZE). I reduced the Y a bit as it is geared down by the belt drive, this quieted travel considerably. I increased the Z a bit as it is being split between the two motors. I found the recommended ~0.55v was about the best for X and E. Tuning was done by ear, with testing at 125mm/s for skipping steps.
     
  11. elmoret

    elmoret Administrator

    Joined:
    Mar 1, 2014
    Messages:
    631
    Likes Received:
    101
    @Miasmictruth it looks to me like stalling, which is when the motion control algorithm can't get data fast enough out of the SD card. If memory serves Simplify3D is particularly bad about this. I think you said you don't experience this with Slic3r, is that true?

    IIRC Simplify3d breaks segments into smaller segments than most slicers.
     
  12. Miasmictruth

    Miasmictruth Well-Known Member

    Joined:
    Sep 4, 2015
    Messages:
    804
    Likes Received:
    118
    Thats partially true it seems less bad with slic3r after further experimenting I was having an issue there too. I need to work on the slic3r profile more as I haven't been able to get it to print right yet I was just running it without filament and watching the motion.

    I did try those numbers for the voltage, I will again tomorrow if I get a chance to work on it, thanks.
     
  13. Miasmictruth

    Miasmictruth Well-Known Member

    Joined:
    Sep 4, 2015
    Messages:
    804
    Likes Received:
    118

    I just had a thought as I am about to go to bed... If I lower the polygon count of the model maybe it will create less gcode when sliced and therefore print smoother... Something to try tommorow.
     
  14. Miasmictruth

    Miasmictruth Well-Known Member

    Joined:
    Sep 4, 2015
    Messages:
    804
    Likes Received:
    118
    So I think @elmoret you are right I saved the model as low instead of high resolution on fusion360, resliced the model in S3D and right away I could see a drop in the number of lines per layer probably about a 10-20 percent difference.

    I extracted 1 layer and stripped away the Z like before and made a script with the before and after with both the new and old code.

    The first 4 passes are the low resolution code at 55mm/s, the second 4 passes are the high resolution code at 55mm/s

    Low res runs smooth in my opinion as soon as it switches to high you can hear/see the chatter as it stalls.



    Slicing past 55mm/s and I get in the same trouble with the low res. I will see what slic3r makes of it tonight, aside from that what can we do to prevent this? A way maybe to force the print to slow down if there if the line segments are too short but alow it to speed up when they are long?

    I am sure if can happen like this it could happen in the middle of other prints depending on geometry.
     
  15. elmoret

    elmoret Administrator

    Joined:
    Mar 1, 2014
    Messages:
    631
    Likes Received:
    101
    It looks as though you're running into a limitation of 8-bit electronics. You can likely tweak with slicer settings and whatnot to mitigate it, but 32 bit electronics (Duet, for example) are better equipped to cope with high polygon/low segment length models.
     
  16. Miasmictruth

    Miasmictruth Well-Known Member

    Joined:
    Sep 4, 2015
    Messages:
    804
    Likes Received:
    118

    Thanks for you help, maybe another future upgrade it sounds like...
     
  17. Miasmictruth

    Miasmictruth Well-Known Member

    Joined:
    Sep 4, 2015
    Messages:
    804
    Likes Received:
    118
    @elmoret I think there is something more going on. My first printer is a cheapo Chinese printer off ebay, I can run the same script and hit around 75mm/s and it still seems smooth as silk. Now while its true that the lower res script preformed better on both printers there no reason I can think of that the BigBox should under preform a 400 dollar printer.

    Could someone run the following scripts on their BigBox and tell me how fast they can run before the motion starts to go wrong. The first script is a low resolution circle I can hit around 55mm/s (less while I am printing for some reason maybe becuse there is no extruding or Z commands) The second script is a high resolution circle I am running both of Repetier host. Both are set to 50mm/s currently

    Script 1
    G1 X115.065 Y85.938 F3000
    G1 X115.558 Y85.432
    G1 X116.590 Y84.476
    G1 X117.132 Y84.024
    G1 X118.257 Y83.180
    G1 X118.843 Y82.786
    G1 X120.049 Y82.062
    G1 X120.672 Y81.730
    G1 X121.946 Y81.133
    G1 X122.600 Y80.868
    G1 X123.929 Y80.405
    G1 X124.606 Y80.208
    G1 X125.976 Y79.885
    G1 X126.670 Y79.758
    G1 X128.065 Y79.578
    G1 X128.769 Y79.523
    G1 X130.175 Y79.487
    G1 X130.881 Y79.505
    G1 X132.283 Y79.614
    G1 X132.984 Y79.705
    G1 X134.367 Y79.957
    G1 X135.055 Y80.119
    G1 X136.405 Y80.512
    G1 X137.072 Y80.744
    G1 X138.375 Y81.274
    G1 X139.014 Y81.573
    G1 X140.256 Y82.234
    G1 X140.861 Y82.598
    G1 X142.028 Y83.383
    G1 X142.593 Y83.807
    G1 X143.673 Y84.708
    G1 X144.191 Y85.187
    G1 X145.173 Y86.195
    G1 X145.639 Y86.725
    G1 X146.512 Y87.828
    G1 X146.921 Y88.403
    G1 X147.676 Y89.590
    G1 X148.024 Y90.205
    G1 X148.653 Y91.463
    G1 X148.936 Y92.110
    G1 X149.432 Y93.426
    G1 X149.646 Y94.099
    G1 X150.005 Y95.459
    G1 X150.149 Y96.150
    G1 X150.366 Y97.540
    G1 X150.438 Y98.242
    G1 X150.511 Y99.647
    G1 X150.511 Y100.353
    G1 X150.438 Y101.758
    G1 X150.366 Y102.460
    G1 X150.149 Y103.850
    G1 X150.005 Y104.541
    G1 X149.646 Y105.901
    G1 X149.432 Y106.574
    G1 X148.936 Y107.890
    G1 X148.653 Y108.537
    G1 X148.024 Y109.795
    G1 X147.676 Y110.410
    G1 X146.921 Y111.597
    G1 X146.512 Y112.172
    G1 X145.639 Y113.275
    G1 X145.173 Y113.805
    G1 X144.191 Y114.813
    G1 X143.673 Y115.292
    G1 X142.593 Y116.193
    G1 X142.028 Y116.617
    G1 X140.861 Y117.402
    G1 X140.256 Y117.766
    G1 X139.014 Y118.427
    G1 X138.375 Y118.726
    G1 X137.072 Y119.256
    G1 X136.405 Y119.488
    G1 X135.055 Y119.881
    G1 X134.367 Y120.043
    G1 X132.984 Y120.295
    G1 X132.283 Y120.386
    G1 X130.881 Y120.495
    G1 X130.175 Y120.513
    G1 X128.769 Y120.477
    G1 X128.065 Y120.422
    G1 X126.670 Y120.242
    G1 X125.976 Y120.115
    G1 X124.606 Y119.792
    G1 X123.929 Y119.595
    G1 X122.600 Y119.132
    G1 X121.946 Y118.866
    G1 X120.672 Y118.270
    G1 X120.049 Y117.938
    G1 X118.843 Y117.214
    G1 X118.257 Y116.820
    G1 X117.132 Y115.976
    G1 X116.590 Y115.524
    G1 X115.558 Y114.568
    G1 X115.065 Y114.062
    G1 X114.136 Y113.006
    G1 X113.698 Y112.452
    G1 X112.883 Y111.306
    G1 X112.504 Y110.710
    G1 X111.811 Y109.486
    G1 X111.496 Y108.854
    G1 X110.932 Y107.565
    G1 X110.683 Y106.905
    G1 X110.256 Y105.565
    G1 X110.076 Y104.882
    G1 X109.788 Y103.505
    G1 X109.679 Y102.808
    G1 X109.535 Y101.408
    G1 X109.498 Y100.703
    G1 X109.498 Y99.297
    G1 X109.535 Y98.592
    G1 X109.679 Y97.192
    G1 X109.788 Y96.495
    G1 X110.076 Y95.118
    G1 X110.256 Y94.435
    G1 X110.683 Y93.095
    G1 X110.932 Y92.435
    G1 X111.496 Y91.145
    G1 X111.811 Y90.514
    G1 X112.504 Y89.290
    G1 X112.883 Y88.694
    G1 X113.698 Y87.548
    G1 X114.136 Y86.994


    Script 2
    G1 X114.227 Y86.874 F3000
    G1 X114.717 Y86.307
    G1 X115.228 Y85.758
    G1 X115.758 Y85.228
    G1 X116.307 Y84.717
    G1 X116.874 Y84.227
    G1 X117.459 Y83.759
    G1 X118.060 Y83.311
    G1 X118.678 Y82.886
    G1 X119.310 Y82.484
    G1 X119.957 Y82.106
    G1 X120.617 Y81.751
    G1 X121.290 Y81.420
    G1 X121.975 Y81.115
    G1 X122.670 Y80.834
    G1 X123.374 Y80.579
    G1 X124.088 Y80.350
    G1 X124.810 Y80.147
    G1 X125.538 Y79.971
    G1 X126.273 Y79.821
    G1 X127.012 Y79.699
    G1 X127.756 Y79.603
    G1 X128.502 Y79.535
    G1 X129.251 Y79.494
    G1 X130.000 Y79.480
    G1 X130.749 Y79.494
    G1 X131.498 Y79.535
    G1 X132.244 Y79.603
    G1 X132.988 Y79.699
    G1 X133.727 Y79.821
    G1 X134.462 Y79.971
    G1 X135.190 Y80.147
    G1 X135.912 Y80.350
    G1 X136.626 Y80.579
    G1 X137.330 Y80.834
    G1 X138.025 Y81.115
    G1 X138.710 Y81.420
    G1 X139.383 Y81.751
    G1 X140.043 Y82.106
    G1 X140.690 Y82.484
    G1 X141.322 Y82.886
    G1 X141.940 Y83.311
    G1 X142.541 Y83.758
    G1 X143.126 Y84.227
    G1 X143.693 Y84.717
    G1 X144.242 Y85.228
    G1 X144.772 Y85.758
    G1 X145.283 Y86.307
    G1 X145.773 Y86.874
    G1 X146.241 Y87.459
    G1 X146.689 Y88.060
    G1 X147.113 Y88.678
    G1 X147.516 Y89.310
    G1 X147.894 Y89.957
    G1 X148.249 Y90.617
    G1 X148.580 Y91.290
    G1 X148.885 Y91.975
    G1 X149.166 Y92.670
    G1 X149.421 Y93.375
    G1 X149.650 Y94.088
    G1 X149.853 Y94.810
    G1 X150.029 Y95.538
    G1 X150.179 Y96.273
    G1 X150.301 Y97.012
    G1 X150.397 Y97.756
    G1 X150.465 Y98.502
    G1 X150.506 Y99.251
    G1 X150.520 Y100.000
    G1 X150.506 Y100.749
    G1 X150.465 Y101.498
    G1 X150.397 Y102.244
    G1 X150.301 Y102.988
    G1 X150.179 Y103.727
    G1 X150.029 Y104.462
    G1 X149.853 Y105.190
    G1 X149.650 Y105.912
    G1 X149.421 Y106.625
    G1 X149.166 Y107.330
    G1 X148.885 Y108.025
    G1 X148.580 Y108.710
    G1 X148.249 Y109.383
    G1 X147.894 Y110.043
    G1 X147.516 Y110.690
    G1 X147.113 Y111.322
    G1 X146.689 Y111.940
    G1 X146.241 Y112.541
    G1 X145.773 Y113.126
    G1 X145.283 Y113.693
    G1 X144.772 Y114.242
    G1 X144.242 Y114.772
    G1 X143.693 Y115.283
    G1 X143.126 Y115.773
    G1 X142.541 Y116.242
    G1 X141.940 Y116.689
    G1 X141.322 Y117.114
    G1 X140.690 Y117.516
    G1 X140.043 Y117.894
    G1 X139.383 Y118.249
    G1 X138.710 Y118.580
    G1 X138.025 Y118.885
    G1 X137.330 Y119.166
    G1 X136.626 Y119.421
    G1 X135.912 Y119.650
    G1 X135.190 Y119.853
    G1 X134.462 Y120.029
    G1 X133.727 Y120.179
    G1 X132.988 Y120.301
    G1 X132.244 Y120.397
    G1 X131.498 Y120.465
    G1 X130.749 Y120.506
    G1 X130.000 Y120.520
    G1 X129.251 Y120.506
    G1 X128.502 Y120.465
    G1 X127.756 Y120.397
    G1 X127.012 Y120.301
    G1 X126.273 Y120.179
    G1 X125.538 Y120.029
    G1 X124.810 Y119.853
    G1 X124.088 Y119.650
    G1 X123.374 Y119.421
    G1 X122.670 Y119.166
    G1 X121.975 Y118.885
    G1 X121.290 Y118.580
    G1 X120.617 Y118.249
    G1 X119.957 Y117.894
    G1 X119.310 Y117.516
    G1 X118.678 Y117.114
    G1 X118.060 Y116.689
    G1 X117.459 Y116.241
    G1 X116.874 Y115.773
    G1 X116.307 Y115.283
    G1 X115.758 Y114.772
    G1 X115.228 Y114.242
    G1 X114.717 Y113.693
    G1 X114.227 Y113.126
    G1 X113.759 Y112.541
    G1 X113.311 Y111.940
    G1 X112.886 Y111.322
    G1 X112.484 Y110.690
    G1 X112.106 Y110.043
    G1 X111.751 Y109.383
    G1 X111.420 Y108.710
    G1 X111.115 Y108.025
    G1 X110.834 Y107.330
    G1 X110.579 Y106.626
    G1 X110.350 Y105.912
    G1 X110.147 Y105.190
    G1 X109.971 Y104.462
    G1 X109.821 Y103.727
    G1 X109.699 Y102.988
    G1 X109.603 Y102.244
    G1 X109.535 Y101.498
    G1 X109.494 Y100.749
    G1 X109.480 Y100.000
    G1 X109.494 Y99.251
    G1 X109.535 Y98.502
    G1 X109.603 Y97.756
    G1 X109.699 Y97.012
    G1 X109.821 Y96.273
    G1 X109.971 Y95.538
    G1 X110.147 Y94.810
    G1 X110.350 Y94.088
    G1 X110.579 Y93.374
    G1 X110.834 Y92.670
    G1 X111.115 Y91.975
    G1 X111.420 Y91.290
    G1 X111.751 Y90.617
    G1 X112.106 Y89.957
    G1 X112.484 Y89.310
    G1 X112.886 Y88.678
    G1 X113.311 Y88.060
    G1 X113.759 Y87.459
     
  18. Miasmictruth

    Miasmictruth Well-Known Member

    Joined:
    Sep 4, 2015
    Messages:
    804
    Likes Received:
    118
    I have a vast improvement but even printing much slower then I think I should have to 20-35mm/s I am still getting artifacts most notably focused in one quadrant of the print.

    [​IMG] [​IMG]

    Tired embedding the photos this time.. Still have to click on them :/
     
  19. Miasmictruth

    Miasmictruth Well-Known Member

    Joined:
    Sep 4, 2015
    Messages:
    804
    Likes Received:
    118
    Here is something interesting, it seems that when I slice using Slic3r the reason it was different it seems it caps the max velocity of circles at 30mm/s somehow. Which the reason I though Slic3r wasn't working either is that I manually boosted the feed rate with repetier host so of coarse I would end up too fast.

    Which I have to award @elmoret another victory lol, thank you again.

    Also Thank you @Alex9779 for running some interesting experiments with me.

    I am still getting some odd things out of S3D before I bought the BigBox I always used Slic3r so still getting used to S3D
     
  20. R Design

    R Design Well-Known Member

    Joined:
    Oct 7, 2015
    Messages:
    944
    Likes Received:
    190
    @Miasmictruth so the conclusion is that, in fact, more bits DOES make a difference in certain situations?

    I remember when this came up last year the conclusion at the time was that the software had not yet been written to take advantage of 32bit systems so Marlin / Rumba was an excellent choice.

    I must admit that, when I printed some cylinders, I didn't notice any jerkiness, but BB did faithfully reproduce the mesh in plastic: the polygonal nature of the "medium" setting export from Fusion360 was VERY visible (and to touch) and even the "high" setting is perceptible if you turn the part at the right angle to the light (on 200 and 250mm diameter cylinders respectively).
     

Share This Page