Unable to get a proper connection in Cura or in ReplcatorG

[kodaimo 0]

I have just assembled my Ultimaker.

Now I connected it to my computer, executed Cura 12.10. It asked for firmware upgrade to Marlin which I agreed upon The upgrade and all the tests went well. I loaded the robot.stl but at the "prepare print" stage I got "something went wrong" and the console is empty. Clicking "print" does nothing.

So I tried with ReplicatorG. There I get the following errors (in the console):

 

[20] Supports RC[20] Connecting to machine using serial port: /dev/ttyACM0[20] Initializing Serial.[20] Attempting to reset RepRap (pulsing RTS)[20] Supports RC[20] Wrong baud rate or faulty firmware detected.[20] Please select the right machine Driver and try again.[20] Firmware version 0.0 detected; firmware version 0.9 required.[20] Wrong baud rate or faulty firmware detected.[20] Please select the right machine Driver and try again.[20] Firmware version 0.0 detected; firmware version 0.9 required.[20] Unknown: g[20] Wrong baud rate or faulty firmware detected.[20] Please select the right machine Driver and try again.[20] Firmware version 0.0 detected; firmware version 0.9 required.[20] Wrong baud rate or faulty firmware detected.[20] Please select the right machine Driver and try again.[20] Firmware version 0.0 detected; firmware version 0.9 required.[20] Unknown: [20] Unknown: j[20] Wrong baud rate or faulty firmware detected.[20] Please select the right machine Driver and try again.[20] Firmware version 0.0 detected; firmware version 0.9 required.[20] Unknown: [20] Wrong baud rate or faulty firmware detected.[20] Please select the right machine Driver and try again.[20] Firmware version 0.0 detected; firmware version 0.9 required.

 

And that's it. It shows "Connecting to Ultimaker v1.0 (5D firmware) on /dev/ttyACM0" but nothing happens.

Under Windows I also tried a demo version of NetFabb. And it cannot connect at all, no matter whatever I set for baud rate.

The only thing I can hear (and feel) is the z-axis motor trying to move or so for 1s, and then stopping. I checked: it is not blocked or anything like that.

Prior to the described I have fixed the rxtx libs bug, chmod'ed /var/lock, /dev/tty* and added myself to dialout group.

I have tested it on some other computer with the same results.

I have no other ideas what to try next. The only hypothesis is that the firmware is buggy, but I don't know how to revert it to the original one. Please help...

[kodaimo 0]

EDIT:

- ReplicatorG was in the most recent version 0037.

- Cura gives the following errors to the console:

 

$ ./Cura-12.10-linux/cura.shException in thread Thread-8:Traceback (most recent call last):File "/usr/lib/python2.7/threading.py", line 551, in __bootstrap_inner   self.run()File "/usr/lib/python2.7/threading.py", line 504, in run   self.__target(*self.__args, **self.__kwargs)File "/home/kodaimo/Cura-12.10-linux/Cura/util/machineCom.py", line 330, in _monitor   self._callback.mcMessage(line)File "/home/kodaimo/Cura-12.10-linux/Cura/gui/configWizard.py", line 424, in mcMessage   self.comm.sendCommand('M119')File "/home/kodaimo/Cura-12.10-linux/Cura/util/machineCom.py", line 508, in sendCommand   self._sendCommand(cmd)File "/home/kodaimo/Cura-12.10-linux/Cura/util/machineCom.py", line 471, in _sendCommand   self._log("Unexpected error while writing serial port: %s" % (getExceptionString()))File "/home/kodaimo/Cura-12.10-linux/Cura/util/machineCom.py", line 405, in _log   self._callback.mcLog(message)File "/usr/lib/python2.7/site-packages/wx-2.8-gtk2-unicode/wx/_core.py", line 14606, in __getattr__   raise PyDeadObjectError(self.attrStr % self._name)PyDeadObjectError: The C++ part of the UltimakerCheckupPage object has been deleted, attribute access no longer allowed.

 

[Daid 306]

 

    self._log("Unexpected error while writing serial port: %s" % (getExceptionString()))

 

That's the interresting part. You're getting an exception the software is trying to write to the serial port. Maybe your USB is flaky. Does dmesg tell you anything?

[kodaimo 0]

As mentioned, I tested it on 2 computers, 3 USB ports in total. The only interesting lines from dmesg are:

 

usb 1-6.3: new full-speed USB device number 27 using ehci_hcdusb 1-6.3: New USB device found, idVendor=2341, idProduct=0042usb 1-6.3: New USB device strings: Mfr=1, Product=2, SerialNumber=220usb 1-6.3: Manufacturer: Arduino (www.arduino.cc)usb 1-6.3: SerialNumber: xxxxxxxxxxxxxcdc_acm 1-6.3:1.0: ttyACM0: USB ACM device... LOTS OF UNRELATED LINES ...usb 1-6.3: USB disconnect, device number 27

 

And I guess they are just fine... No errors.

[Daid 306]

There could be something wrong with your Arduino. You say the initial tests worked, which means it can connect to your machine (because the printing interface is using the same code to connect as the initial tests)

Also, the "something went wrong" is usually the Linux64bit bug, which means you have to replace the pypy in Cura with 64bit pypy. (Linux is the underdog in Cura support at the moment)

[kodaimo 0]

Both the computers were running 32 bit Fedora 17, so it is must be something else. I have no system-wide pypy installed. I also ran Cura under WinVista 32 and same results (apart from error messages, which I didn't look at). The drivers were installed without problems.

How can I test if my Arduino board is faulty or not?

[kodaimo 0]

Actually, I checked which python is executed when I run cura.sh. It is the system-wide one. The pypy is not referenced anywhere in the scripts.

cura.sh:

 

#!/bin/bashpython -c 'import OpenGL'// many lines hereSCRIPT_DIR=`dirname $0`python ${SCRIPT_DIR}/Cura/cura.py $@

 

Cura/cura.py

 

#!/usr/bin/python// ...

 

Trying to run pypy from pypy directory or to run pypy shipped with fedora didn't work because of missing modules or files.

So I grabbed some ancient WinXP 32 installation disk and installed it on the same computer. Then installed python, Cura-12.11, plugged the machine and it worked like from the first time as if there were never any problems with it. Same with an older version of ReplicatorG (0026): worked from the first time.

So I am curious why does it only partially work under WinVista, where it is possible to flash and make tests but nothing more? And why running it under Linux is so problematic? Frankly, before starting to read the Software Setup section on the wiki I was pretty convinced there will be only Linux software. I didn't expect a community which builds their own open source machines, designs they own prints to use anything else...

[Daid 306]

Actually, I checked which python is executed when I run cura.sh. It is the system-wide one. The pypy is not referenced anywhere in the scripts.

Cura/utils/sliceRun.py references and starts pypy.