Windell Oskay

Forum Replies Created

Viewing 15 posts - 1,276 through 1,290 (of 1,518 total)
  • Author
    Posts
  • in reply to: No Communication #21231
    Windell Oskay
    Keymaster

    First, which Linux distribution/version are you using? We developed the Eggbot under Ubuntu, and we’ve had scattered reports of success with other distributions as well. (And, one person reported having trouble with Gentoo last year, but we’ve been unable to reproduce the setup locally.)

    Second, check to see if the Eggbot has shown up as a USB-ACM interface on your computer. The EiBotBoard (EBB) normally shows up on linux machines as /dev/ttyACM* (e.g., /dev/ttyACM0), so look for it there.

    If it is present, try again from Inkscape, at least twice in a row, to see if you can get a connection.

    You may also want to consider unplugging it and plugging it back in, trying a different USB cable (if you have one handy), or if you are using an extra long/thin USB cable or hub, going back to the original cable for testing.

    We have seen some suggestion that USB devices may not register on linux until after a restart, but I’d try that only as a last resort.

    Please let us know how it goes!

    in reply to: eggbot & inkscape – how does one clip or mask? #21230
    Windell Oskay
    Keymaster

    Good news/bad news: It looks like this is a known bug in Inkscape, with a fix committed.  However, the fix is committed for version 0.49, which is not yet released.  Darn.

    in reply to: eggbot & inkscape – how does one clip or mask? #21228
    Windell Oskay
    Keymaster

    On second thought, looking at your picture, it looks like something has gone wrong here.  The right portion of the uppermost straight line has disappeared, and it seems to do this on my computer as well. Yikes.    

    in reply to: eggbot & inkscape – how does one clip or mask? #21226
    Windell Oskay
    Keymaster

    I’ve just tested this again, with the same example, and it seems to work fine. You do still have to click on each of those horizontal line segments, and delete them individually.

    If it’s not working, perhaps the paths were not combined successfully.  If those paths have been combined, then you should be able to click on the combined path (once) and then be able to click and drag it.  Also, when it is selected, the status bar at the bottom of the screen should read “Path (84 nodes) in Layer 1.” 
    in reply to: eggbot & inkscape – how does one clip or mask? #21224
    Windell Oskay
    Keymaster

    >I tried your first method, and it works great for the spiral lines, but not so great for the straight lines. I’ll work on it a bit and see if I can figure out how to do that.

    This technique *should* work equally well for straight lines, and I’ve confirmed (with the example file above) that it does, provided (1) that the oval actually overlaps the straight lines and (2) that you combine (Path>Combine) all of the paths that you wish to cut before selecting them for the Cut Path operation.  You can also use multiple cut path operations, with multiple copies of the oval, to cut the additional paths if you do not wish to combine them.
    in reply to: eggbot & inkscape – how does one clip or mask? #21222
    Windell Oskay
    Keymaster

    Hi William, 

    You’ve brought up a number of issues here, which I’ll try to address one by one.

    First, as the Eggbot is fundamentally a pen plotter, it strives to accurately plot the paths in a drawing. It relies on humans to change the pens and figure out which layer should be drawn in which color. This can lead to some non-WYSIWYG behavior, in the sense that we don’t normally *use* white ink with the Eggbot, but it’s easy to draw white ink on the screen.

    Clipping and masking in Inkscape– and as far as I understand in Illustrator and Corel Draw as well –does not modify the underlying vector drawing, but rather changes which part of the vector drawing is rendered to a bitmap display device: either your screen or an exported bitmap image.   That is to say, no vector drawing of the actual clipped shape is ever created, which is why there is no straightforward way for the Eggbot to draw that clipped shape.    
    For reference, here are a couple of tutorials on clipping and masking in Inkscape and Illustrator:

    http://inkscapetutorials.wordpress.com/2011/02/22/inkscape-faq-how-do-i-crop-in-inkscape/

    Now, I understand that you see serious shortfalls in Inkscape, but so far as I can see, this is fundamentally correct behavior, and in fact the same behavior as you would see in Illustrator or Corel Draw.  It just turns out that this method isn’t really helpful for making Eggbot artwork.   In general, I personally prefer Inkscape, and have on more than one occasion transferred files so that I could edit them in Inkscape before returning them to Illustrator or Corel.  So, I suspect that this is mostly a matter of personal preference.
    To actually create a vector shape that represents your “clipped” vector, there are a few different approaches, depending on what you want to do.  In Illustrator, there are the “pathfinder” operations.  In Inkscape, there are the “boolean” path operations that perform a similar set of actions. 
    In any case, there are actually tools in Inkscape that can cut a complex open-ended path with another: the “Cut Path,” “Difference” and “Division” operations in the Path menu.   Here is an example of how to use it for the “traditional1” pattern.  This is actually a rather unusual example, because we’re trying to cut an open, arbitrary path (rather than a closed object– the more common case) with a second path.
    Here’s the starting figure.
    * Draw an oval that we will punch out.  
    * Duplicate the oval (Edit>Duplicate) to make a second copy, because the oval will be consumed by the Cut Path operation.
    * Click on the oval and lower it to the bottom of the drawing. (Object>Lower to bottom)
    * Select the background path (that which you wish to cut) and the oval.  (Shift-click to select more than one item.)
    * Perform the cut path operation:  Path menu> Cut Path
    The Cut Path operation consumes the selected oval, and cuts the background path everywhere that it intersected the oval.  The oval still visible is the one that we created and placed there by using Duplicate and Lower to bottom.  
    Select the path bits inside the green oval and delete them.  You can select more than one at a time (using Shift-click) to delete a bunch at once.
    All of the inner bits have been removed.  
     
    Select and delete the oval.   Done!
    Also, here is an easier, but “less exact” method.  This method will result in tracing all of the background “propeller” pattern twice, *but* it will print surprisingly quickly because there are fewer pen lifts:
    Begin with the same “traditional1” example pattern, and an oval on top that we wish to punch out.  
    Optional step, for skinnier lines: Select the background “propeller” pattern, and change its line width to 0.001″
    With the background pattern still selected, choose from the menu “Path>Stroke to path.”  This changes that object from being a single line (defined by its stroke only) to being a closed, filled path that is the width of the original line.  
    Now that we have two filled objects to work with, we can use the more common boolean tools.  Select both the oval and the background pattern, and select from the Path menu> Difference.
    And we’re done.
    in reply to: eggbot & inkscape – how does one clip or mask? #21220
    Windell Oskay
    Keymaster

    Have you been unable to do this with the boolean operations or the cut path operation?   I’d suggest that you actually try to modify the path in the way that you want, rather than just try to hide part of it.  

    Windell Oskay
    Keymaster

    Great, glad to hear it!

    Windell Oskay
    Keymaster
    Have you tried adjusting the motor current yet?

    >It acts like they want to move, but more or less have the same pulse that the pen lift motor does.

    Again, can you please describe what kind of a pulse this is?
    Windell Oskay
    Keymaster

    Hi fire-497,

      When you enable the motors, do the steppers seem to actually energize– that is, do they resist motion?  Or, is there actually no apparent change?  Do the steppers seem to move at all when you try to plot?
      And, can you describe what you mean by the “pulses?”
    in reply to: After a certain column, LEDs unresponsive #20891
    Windell Oskay
    Keymaster

    I’m not sure that I understand the situation fully; perhaps you can clarify a few things:

    1.  Columns 0-16 are dim, and 17-25 are bright, even after swapping U4/U5.  Correct?
    2.  Do you have some other LEDs that are not lighting correctly?  
    3.  Have you finished installing the LEDs yet?
    in reply to: FTDI USB – TTL Cable without the FTDI USB? #21212
    Windell Oskay
    Keymaster

    I’m not aware of any standard cable like that, although we have used one or the other end of one of the 6-pin rainbow cables for similar applications now and then:  http://shop.evilmadscientist.com/productsmenu/partsmenu/586

    These cables are pretty easy to cut up and modify, as we demonstrated with the GPS project, for example:
    in reply to: Alpha Clock Five and serial in Python from a Raspberry Pi #21210
    Windell Oskay
    Keymaster

    Okay, so there must be something still screwy about the Pi’s USB driver.  Good to know for future reference, at least. 

    in reply to: Alpha Clock Five and serial in Python from a Raspberry Pi #21208
    Windell Oskay
    Keymaster

    If it is resetting, then somehow you must be triggering the Auto-Reset feature, that we use in programming.  So, you need to disable the RTS# line on the FTDI interface.

    The solution, I believe, is found here: 
    Explicitly disable the RTS as follows:
    myport=serial.Serial(‘/dev/ttyUSB0’,19200, rtscts=1)

     


    Please let me know if that fixes it!
    in reply to: Meggy Jr Question #21206
    Windell Oskay
    Keymaster

    RB6 does control row 7 of the LED matrix, which (with the sideways matrix) is what we’d normally describe as “the second column.”    

    Backing up a bit, “columns” 1,2, and 3 are driven by transistors Q7, Q6, and Q5, from resistors R7, R6, and R5.  If there’s a solder blog connecting those three lines together, or a break in the circuit board traces, it could cause the problem that you’re seeing.  Look carefully at that area of the board, and see if you can figure out what has actually happened there.  If there’s an accidental solder connection, you do need to remove it.
Viewing 15 posts - 1,276 through 1,290 (of 1,518 total)