Forum Replies Created
-
AuthorPosts
-
Windell OskayKeymaster
Can you please say what hardware you are using?
Windell OskayKeymasterYou may need to install the “VCP” driver from FTDI onto your computer, before it will recognize the cable.
Windell OskayKeymasterCan you tell if the problem is one of software or mechanics? In other words, is it genuinely correlated to how complex the drawing is? (And if so, which software are you using?) Can you clarify exactly what is happening when it “loses location for a second?” You said that you can hear it, so that suggests that you’re hearing cogging, or something similar is happening– please elaborate, so that we can help correct the situation.If it is a software issue: There is a known problem in the previous release of RoboPaint (v 0.6.4) that would cause seemingly random lines to appear when the file was too complex. (See here fore more info: https://github.com/evil-mad/robopaint/issues/109 ). If this looks like what you are seeing, and you are using RoboPaint, please make sure that you’re using the latest version (0.7.0).If it is a mechanical issue: Once the WaterColorBot is properly tuned, it will not lose steps anywhere. (And once it’s tuned, it should *stay that way* for months on end.) So, check the steps that we listed above, and make sure that the motor current is appropriate.For what is “appropriate,” please read the link above about adjusting the motor current– we have a lot of hints and advice there. If the current is too low, the motors will be weak, and you’ll get cogging, resulting in a loss of position. (It will not burn out the motor.) If the current is too high, the motors will be hot, jerky, and noisy. There’s always a “goldilocks zone” in the middle.The method of re-calibrating in the middle of painting depends on which software you’re using. For example, in RoboPaint RT, pause the painting, click Motors:Off (to turn off the motors), manually move the carriage to the home position, click Motors:Zero (to declare that the carriage is in the home position), and resume.Windell OskayKeymasterWe have already used the Raspberry Pi method to drive the WaterColorBot, and as compared to the other methods that you’ve mentioned, I would expect it to be one of the simpler.
– For some demos, including at shows and at the White House Science Fair, we had the WCB hooked up to the RPi, running RoboPaint with its API, and receiving data from an iPad. This solution is complex in a way, but also elegant in a sense, since there’s very little at the WCB itself, but you could be sitting with your Chromebook in the back of the room and printing. We could probably also make a change to allow RoboPaint to open a given file and begin printing upon launch.– We’ve also seen people run the Eggbot, with its Inkscape-based driver right from the RPi. (For example, here: http://www.evilmadscientist.com/2013/eggbot-ipad-rpi/ ) It would be just as straightforward to use the WaterColorBot’s Inkscape driver, perhaps with a script that automatically launched the files, and started printing them.– In addition to these options, it is also technically possible to save a program on the EBB itself, for playback. While we have done this in some prior iterations of our firmware, it is not currently implemented, and the available space is rather small. This is likely not one of the best ways to do this, but I though that I’d add it to your list.Windell OskayKeymasterOne note: We have *not* succeeded in using the reservoir extender with our version; it just takes too much power to keep that wax melted all the way up the column.
Windell OskayKeymasterNeat– Thanks for writing this up!
We have also been working on an “official” kistka attachment for the Eggbot, which connects to the EBB, much like the engraver attachment does. That’s certainly possible, but you do need to watch the total power consumption. Our prototypes have been using 0.9 A at 5 V (4.5 W), but it appears that it could be done with considerably less power.Windell OskayKeymasterI don’t see any straightforward way to do this with the same basic architecture, if it has to work with two-pin LEDs. There are, of course, ways to do it, but it would require a very different design, since both the row and column drivers assume that one side is high, and one is low.
If we were to build a Peggy variant that took discrete bicolor LEDs, we would design it so that it took common-anode 3-pin bicolor LEDs. That would be a very straightforward modification.Windell OskayKeymasterDoes the servo motor make any sound, at any point, for example when you launch RoboPaint?Please make sure that the servo connector is connected to the correct port of the EBB, with the black conductor towards the outside of the EBB. If it is not moving at all, ever, then one thing that you might try is reversing the cable, put the white connector towards the outside of the board. (Your carriage and servo motor were tested and calibrated before they shipped, but it is theoretically possible that it somehow passed testing with the cable backwards.)Windell OskayKeymaster@orenjacob I see that you posted this to our github issue report as well. It would be helpful if you would pick one place so that we do not have to have the same conversation twice.
On the RoboPaint RT issue:Can you please say which version of RoboPaint RT you are using? Are you certain that it is the “RoboPaint.RT.Mac-20140101.zip” version? And, which OS version are you on?On the servo issue:In this report, you said that the servo moves freely by hand. In the other report, you said that it seems pretty stiff. Would you please clarify what the situation is?Windell OskayKeymaster@dfwdennis: We have posted a new version of RoboPaint RT that seems to fix this issue (so far as we know– we have still not been able to reproduce it locally), available from our list of releases, under the name “RoboPaint.RT.Mac-20140101.zip”.
Windell OskayKeymasterGreat– Thank you for confirming the fix, and again, I apologize for not having patched our installer yet. I should be able to get to that shortly.
Windell OskayKeymasterHi Mike,
We’ve just had a report of this from another user, and I apologize that we have not had a chance to update the installer file yet.We believe that the issue is due to changes made from version 0.48.2 to version 0.48.4 of Inkscape, so you can probably avoid the error by reverting to that version.Alternately, download and (manually) install this updated version of our extension: https://code.google.com/p/eggbotcode/downloads/detail?name=eggbot_2013_08_11_2.3.4.zip To manually install the extensions, copy the contents of the ZIP archive to (Program files)InkscapeshareextensionsJanuary 4, 2014 at 4:34 am in reply to: Y-Axis not working w. stuttering sound and not moving smoothly #21643Windell OskayKeymasterAwesome– I’m glad that it turned out to be straightforward. :D
January 3, 2014 at 5:25 pm in reply to: Y-Axis not working w. stuttering sound and not moving smoothly #21641Windell OskayKeymasterPlease follow along with the advice in the other thread, here, about how to double check those mechanical parts, and adjust the current if necessary:
If the tension seems correct, and things seem to slide smoothly by hand, it does seem likely to be a current limit adjustment that is needed. However, the fact that it changes as it moves suggests that there may be too much slack in one or both of the cords; you should double check that the rod ends are square at both ends of travel.Windell OskayKeymasterHi Sha,
Fantastic– very glad to hear it!When you get a chance to try a test print, I’d suggest running your first one without a paintbrush inserted– make sure that it looks like it’s doing everything right before using up any more paint and paper, and please let us know how it goes.As it says in the getting started guide, the sandpaper isn’t called for in the instructions, but is there in case you find any rough edges on the WaterColorBot and don’t have any sandpaper hand. ;) -
AuthorPosts