Close



Page 2 of 76 FirstFirst 12341252 ... LastLast
Results 11 to 20 of 757
  1. #11
    Trying out your new code - having the same problem as the std G29 - I have a custom delta with a servo deployed probe. If I run G29 w/switch triggered it moves the head off between towers X & Y) and with switch triggered wants to crash in tower Z. I just can't figure out why it's doing this. Any ideas?

  2. #12
    Super Moderator Roxy's Avatar
    Join Date
    Apr 2014
    Location
    Lone Star State
    Posts
    2,182
    Quote Originally Posted by MarcoM View Post
    Trying out your new code - having the same problem as the std G29 - I have a custom delta with a servo deployed probe. If I run G29 w/switch triggered it moves the head off between towers X & Y) and with switch triggered wants to crash in tower Z. I just can't figure out why it's doing this. Any ideas?
    So I don't know for sure... But in the non-Delta printers there is a case where the nozzle doesn't retract and drags along the bed to go to where it needs to be. There is another thread in this section that RegPye started where we manually fixed the problem with several lines of code. It is a long thread, but how about we do this. How about you review that thread, and then start a new thread detailing your problem (maybe with a short video???) and with some back and forth, we can work through it.

  3. #13
    Super Moderator RobH2's Avatar
    Join Date
    Nov 2013
    Location
    Baltimore, MD
    Posts
    896
    Add RobH2 on Thingiverse
    This is interesting but I can't get it to compile once I change:

    SERIAL_PROTOCOL(matrix[count]); to SERIAL_PROTOCOL(matrix[count],6);

    Here is the error:
    vector_3.cpp:157:35: error: macro "SERIAL_PROTOCOL" passed 2 arguments, but takes just 1
    vector_3.cpp: In member function 'void matrix_3x3::debug(char*)':
    vector_3.cpp:157: error: 'SERIAL_PROTOCOL' was not declared in this scope

    EDIT: If fixed it. I missed the '_F' just before '(matrix[...)'
    Last edited by RobH2; 06-19-2014 at 03:20 PM.
    Bambu P1S/AMS
    NVision4D http://nvision4d.com

  4. #14
    Super Moderator RobH2's Avatar
    Join Date
    Nov 2013
    Location
    Baltimore, MD
    Posts
    896
    Add RobH2 on Thingiverse
    Just ran G29 Tn5 successfully.

    My bed was not too bad. All measurements went from '+' to '-' with none reverting. They all went steadily down. The largest deviation was 1.06 in one but all the other were within .62

    Thanks for this.
    Bambu P1S/AMS
    NVision4D http://nvision4d.com

  5. #15
    Super Moderator Roxy's Avatar
    Join Date
    Apr 2014
    Location
    Lone Star State
    Posts
    2,182
    Quote Originally Posted by RobH2 View Post
    This is interesting but I can't get it to compile once I change:

    SERIAL_PROTOCOL(matrix[count]); to SERIAL_PROTOCOL(matrix[count],6);

    Here is the error:
    vector_3.cpp:157:35: error: macro "SERIAL_PROTOCOL" passed 2 arguments, but takes just 1
    vector_3.cpp: In member function 'void matrix_3x3::debug(char*)':
    vector_3.cpp:157: error: 'SERIAL_PROTOCOL' was not declared in this scope

    EDIT: I fixed it. I missed the '_F' just before '(matrix[...)'

    Good! I was worried the new code base broke the patch when I saw your message!

  6. #16
    Super Moderator DrLuigi's Avatar
    Join Date
    Oct 2013
    Location
    Belgium
    Posts
    502
    Hey Roxy,
    So in the end this Auto leveling is it more precise as mapping the bed ?

    As with the current marlin well, i still see sometimes that it doesnt print evenly,


    Thinking about installing this one atm..

    Also can i still change the depth between glass and nozzle on my Marlin screen at Control or is that gone once i installed your lvling script?

  7. #17
    Super Moderator Roxy's Avatar
    Join Date
    Apr 2014
    Location
    Lone Star State
    Posts
    2,182
    Quote Originally Posted by DrLuigi View Post
    Hey Roxy,
    So in the end this Auto leveling is it more precise as mapping the bed ?
    It is a little hard to answer your question the way it is worded. The upgrade just provides additional functionality, but is intended to be fully compatible with the existing code base.

    By doing a G29 n 5 you will get increased accuracy because any error's in your Z-Probe measurements will average out. Also, you should have:

    #define AUTO_BED_LEVELING_GRID

    turned on because the 3 Point bed leveling (that is still in the code) is not very accurate as you move towards the unprobed corner of the bed.

    Quote Originally Posted by DrLuigi View Post
    As with the current marlin well, i still see sometimes that it doesnt print evenly,
    Well... For sure you can get perfectly even prints. I suspect you are not fully dialed in yet. There are going to be two areas that might need attention.

    First, do a G29 n 5 T and use that to iteratively (manually) level the bed. You need to have the enhanced G29 code to do this. If you don't have it loaded and running, you might want to jump ahead to the second thing. But it is much easier to be accurate with the second thing if you have this part right.

    Look at the Bed Height Topology map that gets printed out. (You can do a G29 n 2 T for the first couple of passes until you get close to speed things up.) It really is helpful to get the first digit after the decimal place to be '0' for each corner. As in +0.0xxx or --0.0xxx You might not be able to get things that 'flat' and the bed leveling will compensate for that. But the closer you start to level, the better off you are. You can find video's of people raising a corner of the bed by 15mm and everything works just fine. Probably, they aren't showing the first video they took. Most likely it took them a couple of tries.

    The other (second) thing you want to check is if you really have your Z-Probe-Offset correct. Once your bed is pretty level, take a look at the bead that gets laid down as the printer is putting down the first layer. You will be able to impact how 'even' the first layer goes down pretty easily by dialing in a 'correct' Z-Probe-Offset for your setup. (And what you measure it to be is just a starting point. The problem is, it is really hard to measure exactly where the micro-switch toggles. So just use your manual measurements as a starting point. Figure out which direction it needs to move and move it a little bit at a time until you get the bead looking how you want it.) The direction it needs to move is sort of inverted. If the nozzle is too close to the bed, you want to make your

    #define Z_PROBE_OFFSET_FROM_EXTRUDER -x.xx closer to 0.0


    After you get both of these items dialed in... I suspect you are going to be saying "This is great!!!!"

    Quote Originally Posted by DrLuigi View Post
    Thinking about installing this one atm..

    Also can i still change the depth between glass and nozzle on my Marlin screen at Control or is that gone once i installed your lvling script?
    I don't understand your question. Marlin doesn't have a screen? Unless you are talking about having an LCD panel in use with Marlin??? If that is what you mean... Yes, you should be able to swap out different plates of different thicknesses and everything should just auto adjust. This code just provides extra functionality to the released G29 code. It doesn't take away any features.
    Last edited by Roxy; 06-20-2014 at 12:29 PM.

  8. #18
    Hi I was hoping someone could help me, ive just tried to change a new version of marlin_main.cpp with the new code youve supplied but i cant get it to compile in arduino. Keeps pointing to the static float probe_pt saying too few arguments to function. I must have done something wrong but cant see what it is.
    Any help is much appreciated.

  9. #19
    Super Moderator Roxy's Avatar
    Join Date
    Apr 2014
    Location
    Lone Star State
    Posts
    2,182
    Quote Originally Posted by brainscan View Post
    Hi I was hoping someone could help me, ive just tried to change a new version of marlin_main.cpp with the new code youve supplied but i cant get it to compile in arduino. Keeps pointing to the static float probe_pt saying too few arguments to function. I must have done something wrong but cant see what it is.
    Any help is much appreciated.
    Not a big deal! Go back and re-read the directions: Pay attention to the part in BOLD: Holler if you have any more problems or questions...


    • At the top of the newly inserted code, enable one of the four lines that corresponds to your printer's configuration. If your origin is at the front left of your bed, you don't have to do anything. That is enabled by default.


    //#define ORIGIN_BACK_LEFT
    //#define ORIGIN_FRONT_RIGHT
    //#define ORIGIN_BACK_RIGHT
    #define ORIGIN_FRONT_LEFT


    • Search for the probe_pt() function. It will look like this: static float probe_pt(float x, float y, float z_before) Delete it. It is roughly 20 lines long. Replace it with the probe_pt.c code attached to this post.



    • Save the changes to Marlin_Main.cpp and pull your Configuration.h file into the text editor.
    Last edited by Roxy; 06-23-2014 at 08:03 PM.

  10. #20
    Yeah done that, deleted and replaced code, origin already at front left. Still says too few arguments. Thank you for the fast response. I'll try and do the whole thing again in the morning, it's 2:25am here.

Page 2 of 76 FirstFirst 12341252 ... LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •