More information during print

[zerspaner_gerd 382]

Hi Tinkergnome,

 

 

Can this be the material exchange speed in your firmware is faster than in the standard Ultimaker 2+ firmware ??

 

The standard UM2.1 firmware (for the "plus" models) uses a fix value of 80 mm/s. So actually the speeds of the tinker fork are a bit slower.

 

I think that was just an illusion!

Have with "Adjust build plate" a comical behavior noted for better description a video

 

If every time when I want to adjust the platform.

Thereafter, every time are my prepared materials away.

Does anybody have the same Problem?

Edited May 6, 2016 by Guest

[Nicolinux 288]

Hey guys,

Sorry for hijacking this thread. I'd like to understand how the UM2 firmware works and thought that if I'am gonna do it, why not start with TinkerGnome's branch and maybe fix something in the process (or break something else badly ).

@TinkerGnome: Where do I find the source code for the UM2+ branch? The repo here contains only the .hex files and two empty source archives:

https://github.com/TinkerGnome/Marlin_UM2plus

And maybe you have a pointer or two about how to setup the toolchain. I am on a Mac (10.11.4).

thx

[tinkergnome 927]

And maybe you have a pointer or two about how to setup the toolchain. I am on a Mac (10.11.4).

I tried to copy the development environment from daid (as far as possible). Try to follow the instructions from these links:

Marlin Simulation... on your desktop?!? (For developers)

Missing note about how to compile and run the simulator

For compiling i use the relevant parts from the (Cura) package.sh (thanks @daid )

This should work on a Mac too (after adjusting some path variables probably).

[gpb01 23]

@tinkergnome: mmm, I think (... but i could be wrong) that Nicolinux needs info (... and me too I would like to know) in which file you define for which kind of Ultimaker2 (std., Go, 2+) you compile the firmware and what are the parameters to modify.

@Nicolinux: me too I'm on OS X 10.11.4 and, for my Ultimaker2 std., I just used the last version of Arduino IDE (1.6.9) to compile the Marlin code that you download from the tinkergnome github (... just tried now, selecting as board "Arduino/Genuino Mega or Mega 2560", compiled Marlin.ino with just one warning )

Hope this help (... still if I have correctly understood the Nicolinux question )

Guglielmo

[derbroti 4]

Some remarks:

first of all the plus firmware is merged with the std one (just different options while compiling) - so use https://github.com/TinkerGnome/Ultimaker2Marlin for everything.

Well "toolchain"... As tinkergnome mentioned in his repo there is a package.sh script which will build the firmware for UM std, +, etc...

On a mac you need to install the Arduino IDE and change the package.sh lines from 43:

 

elif [ -d "/Applications/Arduino.app/Contents/Java" ]; thenARDUINO_PATH=/Applications/Arduino.app/Contents/JavaARDUINO_VERSION=169

 

Adjust the Arduino to the one you are using

You probably need to make package.sh executable (chmod +x package.sh) and you need to create the folders: "resources/firmware/" inside the Marlin folder (mkdir -p resources/firmware).

In the package script you can see the only difference in building the firmware for the different types of Ultimaker2s are the environment variables passed to the make command.

When it is done it produced *.hex files for UM2, UM2+ and UM2dual.

Edited May 13, 2016 by Guest

[Daid 306]

Small note on compiling custom firmwares, I discovered an issue about a week ago, which caused problems for the menu code as soon as the firmware goes over 128k in size (not the size of the hex file, actual binary flash size)

I kinda work around it with this change when you compile with the makefile:

https://github.com/Ultimaker/Ultimaker2Marlin/commit/7408e32de2d4348a077c6b274af3635f022e1e6a

But if you use the Arduino environment, all bets are off now...

[gpb01 23]

Thanks derbroti,

I didn't know the use of "package.sh" ... reading in it is pretty straightforward to derive the parameters to be modified into the source code to compile directly from the Arduino IDE without running any scripts.

Guglielmo

[derbroti 4]

as the firmware goes over 128k in size (not the size of the hex file, actual binary flash size)

 

Shouldn't the avr-gcc linker create jumptable workarounds on his own when reaching the 128k limit for the pointers? (trampoline section) Or is there linker-breaking pointer magic in the lcd code?

Edited May 14, 2016 by Guest

[Nicolinux 288]

Hey guys,

Thanks for the help. Sadly it doesn't work for me - compiling fails due to missing files/dirs.

I am on a Mac and I have two Arduino IDEs installed. Their respective "Java" directories are:

 

• Arduino 1.6.9 - /Applications/Arduino.app/Contents/Java
  
• Arduino 1.0.6 - /Applications/Arduino106.app/Contents/Resources/Java
  

 

 

1. Vanilla UM2Marlin, Arduino 1.6.9

I copied over the package.sh script from @daid. I had to hardcode the Arduino version because the automatic detection doesn't work. After cloning the git sources the script stops with this (sh -x for debug output):

 

+ make -j 3 HARDWARE_MOTHERBOARD=7 ARDUINO_INSTALL_DIR=/Applications/Arduino.app/Contents/Java ARDUINO_VERSION=169 BUILD_DIR=_UltimakerMarlin_250000 'DEFINES='\''VERSION_BASE="Ultimaker:_15.04.6"'\'' '\''VERSION_PROFILE="250000_single"'\'' BAUDRATE=250000 TEMP_SENSOR_1=0 EXTRUDERS=1'make: *** No rule to make target `_UltimakerMarlin_250000/wiring.o', needed by `_UltimakerMarlin_250000/3D'.  Stop.

 

 

2. Vanilla UM2Marlin, Arduino 1.0.6

This time it starts compiling but then stops because of a (presumably hardcoded missing directory "UM2"):

 

+ make -j 3 HARDWARE_MOTHERBOARD=72 ARDUINO_INSTALL_DIR=/Applications/Arduino106.app/Contents/Resources/Java ARDUINO_VERSION=106 BUILD_DIR=_Ultimaker2 'DEFINES='\''STRING_CONFIG_H_AUTHOR="Vers:_15.04.6"'\'' TEMP_SENSOR_1=0 EXTRUDERS=1' CC    /Applications/Arduino106.app/Contents/Resources/Java/hardware/arduino/cores/arduino/wiring.c CC    /Applications/Arduino106.app/Contents/Resources/Java/hardware/arduino/cores/arduino/wiring_analog.c[...] CXX   Servo.cpp CXX   /Applications/Arduino106.app/Contents/Resources/Java/hardware/arduino/cores/arduino/Tone.cpp CXX   /Applications/Arduino106.app/Contents/Resources/Java/hardware/arduino/cores/arduino/main.cpp CXX   UM2 CXX   _Ultimaker2/3D CXX   Marlin.elfavr-gcc: avr-gcc: avr-gcc: UM2: No such file or directoryUM2: No such file or directoryUM2: No such file or directory

 

 

3. Tinkergnome, Arduino 1.6.9

Used Tinkergnome customized package.sh script and commented out the Arduino detection code to force it to take the latest Arduino version.

 

+ make -j 3 HARDWARE_MOTHERBOARD=72 ARDUINO_INSTALL_DIR=/Applications/Arduino.app/Contents/Java ARDUINO_VERSION=169 BUILD_DIR=_Ultimaker2 'DEFINES='\''STRING_CONFIG_H_AUTHOR="Tinker_16.03.1"'\'' TEMP_SENSOR_1=0 EXTRUDERS=1 FILAMENT_SENSOR_PIN=30 BABYSTEPPING HEATER_0_MAXTEMP=315 HEATER_1_MAXTEMP=315 HEATER_2_MAXTEMP=315' CC    /Applications/Arduino.app/Contents/Java/hardware/arduino/avr/cores/arduino/wiring.c[...] CXX   /Applications/Arduino.app/Contents/Java/hardware/arduino/avr/cores/arduino/new.cpp CXX   /Applications/Arduino.app/Contents/Java/hardware/arduino/avr/cores/arduino/main.cpp CXX   UM2 CXX   _Ultimaker2/3D CXX   Marlin.elf COPY  Marlin.hex/bin/sh: line 0: [: too many arguments+ cp _Ultimaker2/Marlin.hex resources/firmware/Tinker-MarlinUltimaker2-16.03.1.hexcp: _Ultimaker2/Marlin.hex: No such file or directory

 

 

4. Tinkergnome, Arduino 1.0.6

Same error as Vanilla UM2Marlin and Arduino 1.6.9

 

+ make -j 3 HARDWARE_MOTHERBOARD=72 ARDUINO_INSTALL_DIR=/Applications/Arduino106.app/Contents/Resources/Java ARDUINO_VERSION=106 BUILD_DIR=_Ultimaker2 'DEFINES='\''STRING_CONFIG_H_AUTHOR="Tinker_16.03.1"'\'' TEMP_SENSOR_1=0 EXTRUDERS=1 FILAMENT_SENSOR_PIN=30 BABYSTEPPING HEATER_0_MAXTEMP=315 HEATER_1_MAXTEMP=315 HEATER_2_MAXTEMP=315'make: *** No rule to make target `_Ultimaker2/wiring.o', needed by `_Ultimaker2/3D'.  Stop.+ cp _Ultimaker2/Marlin.hex resources/firmware/Tinker-MarlinUltimaker2-16.03.1.hexcp: _Ultimaker2/Marlin.hex: No such file or directory

 

 

Reminds me why I didn't want to do this in the first place. Programming is fun but package management and dependencies are not :/

[Nicolinux 288]

Ok, I found it.

I had spaces in my directory structure (yes I know, if you Unix, then that's stupid). My path was:

 

/Users/stefan/Projects/3D Drucker/Ultimaker/UM2 Firmware/nicolinux/Marlin

 

and the $TARGET from the Makefile (created with "TARGET = $(notdir $(CURDIR))") was:

 

3D UM2 Marlin

 

Fixed it. Now the firmware compiles but I get these strange warnings (I'll ignore it since it occurs in "electronics_test" and hope it is just a test that fails):

 

[...] CXX   Servo.cpp CXX   /Applications/Arduino106.app/Contents/Resources/Java/hardware/arduino/cores/arduino/Tone.cpp CXX   /Applications/Arduino106.app/Contents/Resources/Java/hardware/arduino/cores/arduino/main.cpp CXX   _Ultimaker2/Marlin.elftarget: Marlin_Ultimaker2/electronics_test.o:(.progmem.gcc_sw_table+0x2e): warning: internal error: out of range error_Ultimaker2/electronics_test.o:(.progmem.gcc_sw_table+0x30): warning: internal error: out of range error_Ultimaker2/electronics_test.o:(.progmem.gcc_sw_table+0x32): warning: internal error: out of range error_Ultimaker2/electronics_test.o:(.progmem.gcc_sw_table+0x34): warning: internal error: out of range error_Ultimaker2/electronics_test.o:(.progmem.gcc_sw_table+0x36): warning: internal error: out of range error_Ultimaker2/electronics_test.o:(.progmem.gcc_sw_table+0x38): warning: internal error: out of range error_Ultimaker2/electronics_test.o:(.progmem.gcc_sw_table+0x3a): warning: internal error: out of range error_Ultimaker2/electronics_test.o:(.progmem.gcc_sw_table+0x3c): warning: internal error: out of range error_Ultimaker2/electronics_test.o:(.progmem.gcc_sw_table+0x3e): warning: internal error: out of range error_Ultimaker2/electronics_test.o:(.progmem.gcc_sw_table+0x40): warning: internal error: out of range error_Ultimaker2/electronics_test.o:(.progmem.gcc_sw_table+0x42): warning: internal error: out of range error_Ultimaker2/electronics_test.o:(.progmem.gcc_sw_table+0x44): warning: internal error: out of range error_Ultimaker2/electronics_test.o:(.progmem.gcc_sw_table+0x46): warning: internal error: out of range error_Ultimaker2/electronics_test.o:(.progmem.gcc_sw_table+0x48): warning: internal error: out of range error_Ultimaker2/electronics_test.o:(.progmem.gcc_sw_table+0x4a): warning: internal error: out of range error_Ultimaker2/electronics_test.o:(.progmem.gcc_sw_table+0x4c): warning: internal error: out of range error COPY  _Ultimaker2/Marlin.hexAVR Memory Usage----------------Device: atmega2560Program:  134910 bytes (51.5% Full)(.text + .data + .bootloader)Data:       5999 bytes (73.2% Full)(.data + .bss + .noinit)  text	   data	    bss	    dec	    hex	filename134614	    296	   5703	 140613	  22545	_Ultimaker2/Marlin.elf

 

[derbroti 4]

"/Applications/Arduino106.app"

You should use your more recent version 1.6.9 of the Arduino IDE.

(tinkergnome, at least for the last release, is using Arduino 1.6.5, and his builds are good - at least no one is complaining )

 

(I'll ignore it since it occurs in "electronics_test" and hope it is just a test that fails):

 

Please don't, it is bad practice to ignore warnings - often there are harmless but there is a reason that gcc has a switch to treat all warnings as errors

Edited May 15, 2016 by Guest

[Nicolinux 288]

I know, I did that too but it fails due to other reasons. It was one of the error scenarios from my previous post (3. Tinkergnome, Arduino 1.6.9). I saw that Tinker did catch some changes in the directory structure with the new Arduino IDE 1.6.9 but it still fails for me...

[derbroti 4]

Are you sure you did nothing else in 3.?

I just re-downloaded 1.6.9 and tinkergnomes master and with just editing the three lines mentioned (the detection if clause and the two assignment lines below that) and it works fine.

For example in your 1. you had a wrong mainboard define "HARDWARE_MOTHERBOARD=7" instead of 72.

Try a clean restart:

elif [ -d "/Applications/Arduino.app/Contents/Java" ]; then <- edit here to match the line below

       ARDUINO_PATH=/Applications/Arduino.app/Contents/Java <- make sure same path as the if and a valid one

       ARDUINO_VERSION=169 <- version works fine

and then the package.sh will work fine with 1.6.9

Edited May 15, 2016 by Guest

[Nicolinux 288]

About the warnings - you are right, but not ignoring implies understanding the issue - and that I did not. I looked at electronics_test.cpp but since I know very little about the project structure, I couldn't make up what it does other than you know... testing the electronics

(ok, maybe communicating with the hardware is involved too).

This comment in the header file reinforces my feeling that I can safely ignore it:

 

//Start the electronics test environment. This environment never returns and disables all other features.

However, there is still one thing that I don't understand. What is the difference between these two git repos?

https://github.com/Ultimaker/Ultimaker2Marlin

and

https://github.com/Ultimaker/UM2.1-Firmware

Looking at daid's package.sh script, it checks out different repos for different UM2 variants. Tinker's package.sh does not do that but creates firmware versions for different UM2 printers.

I am asking because I have finally compiled my own firmware with a modified version of Tinker's package.sh and the first run wizard just tried to crash the heated bed into the print head because it thought I have an UM2+ Extended (with lightning fast reflexes I switched it off before tears where shed.)

[derbroti 4]

Go and extended(+) have different branches; std/std(+) are in master. That's why the Cura package.sh checks out different branches.

UM2.1 should be the new firmware that will ship with the new Cura 2.1

Edited May 15, 2016 by Guest

[Nicolinux 288]

Alright, thanks. I'll try again (from scratch).

[gpb01 23]

@Nicolinux: ... can I give you a suggestion ?

One very nice possibility of the new IDE (if I remember well is starting from the 1.6.5) is to have "sandboxed" independent IDE ... you can have many versions of the same/different IDE and each one is totally closed i herself.

To do that, on OS X (... where I'm) first give a unique name to your Arduino copy (eg. the last version, for me, is "Arduino 1.6.9 Portable.app"), next right click and select "Show Package Contents", next go to "Contents/Java/"  and inside create a new folder named "portable" (so you have Contents/Java/portable/"). Close and restart the IDE ... you have a sandboxed IDE, totally independent from other IDE installation (so ... NO the same libraries, NO the same packages, NO the same boards, ecc.) very useful to create specific IDE installations.

If you download the IDE 1.6.9, give a unique name, make "portable" (as explained), select "Arduino MEGA 2560", open "Marlin.ino" and compile ... you should have just one warning that you can ignore

Guglielmo

Edited May 16, 2016 by Guest

[Nicolinux 288]

Thanks Guglielmo, that's a good idea. It is very nice that you can do this kind of stuff on a Mac easily.

Oh man, Tinker's firmware is super cool. Not only the display style but also so many convenience features. One can even PID autotune directly from the menu.

I'll need to wrap my head around the code, but for now, looking at it from the user perspective it is very very nice.

@tinkergnome you rock!

[Nicolinux 288]

I assume many of you use OctoPrint too with this firmware. How is it supposed to work?

When I connect with OctoPrint I see the following on the printer's display:

Lights go out and the printer freezes.

OctoPrint shows the following:

I assume OP tries to cache the entire sd card contents and takes quite a while. Sadly I wasn't patient enough to wait for too long, but I did it twice and it "hang" on the same filename in the OP terminal.

I guess that's more of a OctoPrint question than firmware related. So sorry for misusing this poor thread yet again...

[Nicolinux 288]

Hm it seems the weirdness doesn't stop yet. I've had several prints abort with "Error: Tried printing out of printing area".

There are some strange moves that have nothing to do with the object - the print head just moves outwards and returns in a straight line.

[gr5 2,173]

Try changing USB cables or adding a USB hub or using a different computer to drive the USB signal (are you using raspberry pi?)

[Nicolinux 288]

Not using OctoPrint with this firmware yet because of the freezing with "M20" commands.

These were regular prints from the sd card. I swapped the sd card with a brand new one, but it still persists.

[Dim3nsioneer 557]

Interesting error message on the display. Trying to print outside of the valid area reminds me of a very early Cura2 bug (at a time it was called 'Pink unicorn') where movements to almost infinity happened. But please don't ask me about details of that bug....

So, as you are printing from sd card, the wrong gcode should be there... is it?

[derbroti 4]

This looks excatly like my problem: 21252-weird-axis-movement-mid-print

The problem was a broken SD card.

It would explain the out of printarea error as the wrongly read gcode would make the printer move outside the build area.

Can you try a different one as well?

Just read you already swapped it - maybe the new one is just faulty?

Do you have by chance a third one?

Are you using a micro-sd to sd adapter? maybe that one is faulty (bad contacts).

Edited May 29, 2016 by Guest

[Nicolinux 288]

@Dim3nsioneer: I stepped through the .gcode file and didn't found any strange moves outside of the printing area.

@derbroti: I read about possible sd card problems, that's why I took a new one.

Now I went on and just printed the same .gcode file again from the same sd card. And the print finished properly. What do you make of it?