Auto squaring and end stops

New Home Forum Mostly Printed CNC – MPCNC Advice – MPCNC Auto squaring and end stops

This topic contains 96 replies, has 7 voices, and was last updated by  Ryan 1 year ago.

Viewing 30 posts - 61 through 90 (of 97 total)
  • Author
    Posts
  • #46803

    Jeffeb3
    Participant

    Maybe there’s not an easy way to run Travis outside of GitHub. I would be fixing this offline before pushing the fix. But then you couldn’t watch…

    #46805

    Ryan
    Keymaster

    I just looked through the travis site, it seems to be tied to gethub, but not part of it? This last test was so much closer. It didn’t totally blow up just one little error with the Z axis and we didn’t touch that one! High five.

    #46808

    Jeffeb3
    Participant

    I just looked at running Travis locally. That’s clearly not part of their business model, so they haven’t designed for it.

    I did find the funniest name for a package ever: https://github.com/grosser/wwtd

    W.W.T.D. What Would Travis Do? Perfect name.

    It does look like you can build a docker image similar to what Travis uses. But then you’re stuck running each command in the travis.yaml yourself. It would be nice if the configs from the failed builds were available at least, then you could plop them into your local checkout and at least get pretty confident it would build.

    Also, also, I wonder if those config scripts would help you with your deliveries.

    #46809

    Ryan
    Keymaster

    I did see the config scripts wondering how it was happening. I think the software you built for me will work, I seem to understand it well enough. I have just been so sick of looking at the firmware, merging, compiling and testing I was going to wait to use yours on the next round firmware updates. That email is one of the only ones in my inbox…and i’m pretty good at keeping my inbox at zero.

    #47001

    Ryan
    Keymaster

    Where do I adjust how much the z axis raises before homing?

    I am tuning in the new firmware now that it is merged. Can you tell me why you wanted to change it? Was there an issue or just personal preference? To me it seems the safest thing to do is pick the Z axis up before moving.

    #47011

    Andy M.
    Participant

    Well at the time I was still figuring out my g code and how I wanted to use the machine with a home position. The clearence of the Z turned out to be a non issue  after a I figured out what i was doing. It’s a good thing, it would be nice to know what that clearence is. There could be instances where you might want it higher or lower.  You could always do this with g code. The software endstops really messed me up at first, and had me blaming the clearence plane of the Z. But it was just the non ability of going negative in the Z and my correct commands causing it to move in the opposite direction. So, clearence is good on Z but should be notated as to what that clearence is. In my opinion the software end stops should be completely disabled as well in the firmware…all in all after trial and error it’s so nice to have a home position I can rely on. Thanks Ryan. I want to put a video or two up of the feature when I find the time.

     

    Thanks,

    Andy

    #47012

    Jeffeb3
    Participant

    We also got a change in to configure soft stops by axis, so you should be able to leave on in X and Y but turn it off in Z.

    I’m not sure anyone has successfully tested it, but it’s in bugfix.

    #47291

    Ryan
    Keymaster

    I have tried a few times, still nothing. I haven’t looked to see any obvious problems though.

    #47302

    Jeffeb3
    Participant

    Damn. I’ll have to check this out. The code was so simple. Can you PM me a link or zip to exactly what you’re running?

    #47321

    Andy M.
    Participant

    Quick overview video as promised.

    #47322

    Ryan
    Keymaster

    Like the video!

    Dupont 2.54mm (0.1″)

    I think your touch plate interference is from the spindle power supply.

    We can tighten up your Z routine, soon. Sorry it has been a busy week and setting up the new firmware keeps getting pushed as I prep orders. Worked so dam hard to get it merged now I am lagging with instructions….lame.

    #47350

    Andy M.
    Participant

    Thanks Ryan, nice clarification on the pin spacing of the DuPont connectors…Ya theres a good possibility that could be causing it…at this point the paper is working great, so I’ll stick with it for a while…

    Hey man don’t beat yourself up we all get busy and get behind, you can trust me on that! Running a small business isn’t easy…takes commitment, infinite  patience, and a lot of a thing called hard work. My Dad’s been running one for 30 years, and I’ve been right there experiencing it. I have great respect for the ones that make a go of it. Keep up the good work.

     

    Andy

    1 user thanked author for this post.
    #47362

    Jeffeb3
    Participant

    2.54mm is 1/10th” or 0.1″ /pedantic

    I just updated my MP3DP to bugfix 1.1.x, and I tried the softstops by axis configuration, and it worked the way I expected.

    Configuration.h

    // Min software endstops curtail movement below minimum coordinate bounds
    #define MIN_SOFTWARE_ENDSTOPS
    #if ENABLED(MIN_SOFTWARE_ENDSTOPS)
    #define MIN_SOFTWARE_ENDSTOP_X
    #define MIN_SOFTWARE_ENDSTOP_Y
    // #define MIN_SOFTWARE_ENDSTOP_Z
    #endif
    
    // Max software endstops curtail movement above maximum coordinate bounds
    #define MAX_SOFTWARE_ENDSTOPS
    #if ENABLED(MAX_SOFTWARE_ENDSTOPS)
    #define MAX_SOFTWARE_ENDSTOP_X
    #define MAX_SOFTWARE_ENDSTOP_Y
    // #define MAX_SOFTWARE_ENDSTOP_Z
    #endif
    
    

    After that, I had the hotend in the middle of the space somewhere, powered on and through the LCD menu, I tried moving around. I could move negative in the Z, but not in the X or Y. I think that’s what we want. That was before homing, and then I tried it again after homing. I assume 2.0 would work the same.

    On another note, while looking through the configuration, I noticed a comment that points out M120/M121, which will enable/disable the endstops. So if anyone is having probe/interference issues, you can add M121 to the start of your cut, and M120 to the end.

    http://reprap.org/wiki/G-code#M120:_Enable_endstop_detection

    #47367

    Ryan
    Keymaster

    Sweet heffe I am glad to hear it. The hacks I had to do to Arduino to get it to work with the 2.0 version must be the issue. I had another problem with it a few days ago. It doesn’t let go of the compiler data or something I could delete whole lines out of the configs and if the versions didn’t change they’d still compile. I have to reinstall Arduino and try to figure out what the heck is going on. To get things to work I had to use the newest Arduino beta so I knew it didn’t have the hacks in it that seems to work.  Today my plan is to get all the firmwares done and out and do some testing.

    Sorry to have made you get out the ramps. I knew something was fishy that’s why I didn’t make a big stink.

    #47369

    Jeffeb3
    Participant

    I think you might be seeing why people like platform.io 😉

    The Arduino IDE doesn’t reload file changes outside the program unless you have “use external editor” on in the preferences.

    I tested with 1.1. I’m not trusting 2.0 until there’s a release. I hope there’s nothing wrong with the soft endstops in 2.0, but I trust thinkyhead do have integrated that correctly.

    I have ramps on my MP3DP, my piCNC on my low rider, and my mini Rambo on the zen (Which I intend to move to the MP3DP). I also didn’t try dual endstops, just the soft endstops.

    #47373

    Ryan
    Keymaster

    I know its all getting so complicated and dam marlin is moving really fast right now, so many updates and new features. If they start focusing on just one branch we will have some serious innovations.

    I have a stack of boards in front of me that all need firmware and testing….Might be time to set up that script.

    I was just told about what seems like a pretty mature python firmware “kippler” Might be right up your alley?!?

    #47381

    Jeffeb3
    Participant

    Yeah, I was thinking about that script again this morning. I had to make a bunch of identical edits for my MP3DP for the version I was interested in, and I thought that maybe I should just make the script do it. But, it wasn’t bad for me.

    I can’t find any kippler stuff (besides info on the fish). This is CNC firmware?

    Also, sorry Andy, we’ve hijacked your thread.

    #47383

    Ryan
    Keymaster

    My impeccable spelling might have been the cause.

    https://github.com/KevinOConnor/klipper

    #47403

    Ryan
    Keymaster

    Dang Heffe I figured it out…I doesn’t work in repetier host after homing, but it does work with the LCD, I typically only test with repetier since it’s faster….Shoot, I don’t see an easy way around it in the settings either.

    // Min software endstops constrain movement within minimum coordinate bounds
    #define MIN_SOFTWARE_ENDSTOPS
    #if ENABLED(MIN_SOFTWARE_ENDSTOPS)
    #define MIN_SOFTWARE_ENDSTOP_X
    #define MIN_SOFTWARE_ENDSTOP_Y
    //#define MIN_SOFTWARE_ENDSTOP_Z
    #endif

    Andy, I have a 2.0 ramps dual endstop firmware on my github page, when you get a minute can you see if you can compile it. It works for me as is with arduino 1.8.5.

    #47406

    Jeffeb3
    Participant

    Ok. I can try that easily, now that I’ve upgraded to 1.1

    #47418

    Andy M.
    Participant

    Ya sure i can do that when i get a chance. No worries on the hijack, no two better people to hijack a thread lol

    #47445

    Jeffeb3
    Participant

    I did some more detailed testing, and this is what I found.

    If you power on the machine in the middle somewhere, and try to move into the negative space, it will in the Z and not in X, Y. I found that with the LCD or the computer (I was using pronterface, I hope that isn’t making a difference).

    If I go into the positive somewhere, and issue a G92 X0 Y0 Z0, it will then go negative, but not more than the original 0. For example:

    G01 X1 Y1 Z1
    G92 X0 Y0 Z0
    G01 X-1 Y-1 Z-1 ; OK
    G01 Z-2 ; OK, because of that setting
    G01 X-2 Y-2 ; Doesn’t go there.

    I was using the buttons to move around, mostly. I assume that these, and RH’s buttons are both just sending the G01 commands.

    I’m surprised it has any concept of machine coordinates or home coordinates.

    Another thing I noticed is that the “reset” button in pronterface wouldn’t always reset it, but if it showed the vicious logo, then it was actually reset.

    If you are still finding the problem, can you paste or email me a .zip and maybe a gcode file that illustrates the problem? Some gcode snippets are fine too. It’s just more specific than english.

    #47449

    Ryan
    Keymaster

    I think it is fine, no one should be manually moving negative after using a touch plate and it works with gcode so perfecto. The new firmware has it set.

    #47467

    Andy M.
    Participant

    So the bug fix 2.0 is fully functional and the one I want? As Ryan said I don’t care about manually moving negative in x and y as long as I can home all the way at any given point.Goina try this tonight.

    #47469

    Ryan
    Keymaster

    Mine should be, I edited and tested a bunch last night and have been doing it all day so it should be good but you know me, far from perfect the first time, every time.

    #47470

    Andy M.
    Participant

    Ok great I’ll let you know how it works out…here’s a picture of the finished map sign in the video…I think the only other project picture I have posted here was the test map… what’s with me and maps your thinking?…I swear I don’t just set around cutting them all the time lol

    #47472

    Ryan
    Keymaster

    Looks awesome, adding it to the gallery.

    1 user thanked author for this post.
    #47497

    Jeffeb3
    Participant

    That looks great. I think Hawaii and Alaska are in the wrong place though.

    It looks really good though. Just wrong. 🙂

    #47500

    Barry
    Participant

    I’m just glad he got the whole state!  Most people forget to add the sliver that includes the state capitol…

    #47865

    Andy M.
    Participant

    Just now getting around to trying the new firmware. I get the following message when i try to compile it:

    Arduino: 1.8.4 (Windows 7), Board: “Arduino/Genuino Mega or Mega 2560, ATmega2560 (Mega 2560)”

    fork/exec C:\Program Files\Arduino\hardware\tools\avr/bin/avr-gcc.exe: The filename or extension is too long.
    Error compiling for board Arduino/Genuino Mega or Mega 2560.

    This report would have more information with
    “Show verbose output during compilation”
    option enabled in File -> Preferences.

     

    I notice it looks completely different in file explorer from the other firmware i have compiled. Meaning the others where not broke into folders and were just a big list of files with the .ino somewhere in the middle. This new firmware has everything in folders and i am only finding a .ino file under the “marlin” folder. Maybe i did something wrong with github?? I downloaded both of the ones that specify for ramps and duel end stops and tried them. If you show “Verbose output” it instantly hits this file location and keeps looping until it errors out. I’m guessing this is a dumb mistake on my part. I have a feeling its something simple i’m just not familiar with. Is there a step i’m missing before pulling it into Arduino? Thanks for your help.

     

    Andy

Viewing 30 posts - 61 through 90 (of 97 total)

You must be logged in to reply to this topic.