search

Klipper3d / klipper

Klipper is a 3d-printer firmware
GNU General Public License v3.0
9.08k stars 5.21k forks source link

anet a8 #78

Closed manotroll closed 6 years ago

manotroll commented 6 years ago

has support for anet a8 do you have support for automatic leveling? Do you have temperature protection?

KevinOConnor commented 6 years ago

On Thu, Dec 28, 2017 at 08:43:39AM -0800, manotroll wrote:

has support for anet a8 do you have support for automatic leveling? Do you have temperature protection?

I don't know what electronics are in the anet a8, so I can't be sure.

Automatic bed leveling still needs to be implemented (see the docs/Todo.md file).

The micro-controller software will detect if a heater is outside the min_temp/max_temp range and disable the heaters if so. Some more advanced detetion of subtle heater failures (eg, the thermistor falling off the heater) have not yet been implemented.

-Kevin

rewolff commented 6 years ago

FYI, while building two new 3D printers over the last few months, I've had about three times that the marlin firmware detected something odd going on and went into "safe mode". All false positives, but still it is nice to know that it works.

Now I must say that Marlin's handling of that situation is far from ideal: It shuts down completely, does not report temperatures anymore, and sometimes a whole reboot of the octopi machine is necessary to get things going again.

So... learning from that. When such an error condition is detected, keep responding to commands, but make sure that the heaters are in a "safe" mode for a while. Make sure that "manual intervention" happens before you enable them again. So maybe require that the temperature is set to zero before another temperature target can be set?

For context as to when something like this could happen: I have an ATX powersupply that powers my machine. Apparently it has a feedback loop on the 5V and not on the 12V line: When the 3D printer loads the 12V line it drops to 11V and even to 10V when both bed and hotend are on. At 10V my hotend cannot reach the target temperature for my plastic: 225 degrees. A workaround is to load the 5V with one coil of a stepper motor I have.... So when Marlin detects: "I turned on the heater, but it took too long for the temperature to go up", I can reconnect the dummyload and get things to work better. Fault cleared, it would be nice to be able to allow the system to continue without having to reboot.

manotroll commented 6 years ago

I don't know what electronics are in the anet a8, so I can't be sure.

Automatic bed leveling still needs to be implemented (see the docs/Todo.md file).

The micro-controller software will detect if a heater is outside the min_temp/max_temp range and disable the heaters if so. Some more advanced detetion of subtle heater failures (eg, the thermistor falling off the heater) have not yet been implemented.

right as soon as the customs finish reviewing the taxes I will test I already have a pi 3 + b ready the videos I saw on YouTube show an incredible speed I have completely discharged the marlin

FYI, while building two new 3D printers over the last few months, I've had about three times that the marlin firmware detected something odd going on and went into "safe mode". All false positives, but still it is nice to know that it works.

Now I must say that Marlin's handling of that situation is far from ideal: It shuts down completely, does not report temperatures anymore, and sometimes a whole reboot of the octopi machine is necessary to get things going again.

So... learning from that. When such an error condition is detected, keep responding to commands, but make sure that the heaters are in a "safe" mode for a while. Make sure that "manual intervention" happens before you enable them again. So maybe require that the temperature is set to zero before another temperature target can be set?

For context as to when something like this could happen: I have an ATX powersupply that powers my machine. Apparently it has a feedback loop on the 5V and not on the 12V line: When the 3D printer loads the 12V line it drops to 11V and even to 10V when both bed and hotend are on. At 10V my hotend cannot reach the target temperature for my plastic: 225 degrees. A workaround is to load the 5V with one coil of a stepper motor I have.... So when Marlin detects: "I turned on the heater, but it took too long for the temperature to go up", I can reconnect the dummyload and get things to work better. Fault cleared, it would be nice to be able to allow the system to continue without having to reboot.

uses a better source with more amperage or a separate source set to 13.8V

rewolff commented 6 years ago

Yeah, for a little more margin, i ordered a 15V, 40A powersupply. It arrived several minutes after writing the above comment.

manotroll commented 6 years ago

and solved it?

rewolff commented 6 years ago

Solved what?

I explained that I've seen from Marlin. This is intended to improve klipper: The same mistakes should be avoided.

The supply was advertised with "fan makes a little noise". Which could be interpreted as chinglish for "fan makes little noise: very quiet". However, this seems to be "fan is a little noisy" which is an euphemism for "fan is very loud (and always-on)".

C4g2 commented 6 years ago

Just got my Anet A8 via Sain Smart! The mainboards IC is an Atmel branded ATMEGA1284p AU1720

Any other info you need to confirm compatibility? I'm VERY interested if this could run my machine.

manotroll commented 6 years ago

Solved what?

I explained that I've seen from Marlin. This is intended to improve klipper: The same mistakes should be avoided.

The supply was advertised with "fan makes a little noise". Which could be interpreted as chinglish for "fan makes little noise: very quiet". However, this seems to be "fan is a little noisy" which is an euphemism for "fan is very loud (and always-on)".

you did not have a drop problem for 10V and the 15V source arrived if you solved your reset problem

Just got my Anet A8 via Sain Smart! The mainboards IC is an Atmel branded ATMEGA1284p AU1720

Any other info you need to confirm compatibility? I'm VERY interested if this could run my machine.

C4g2 I think trying the same ramps settings might be a good way to start perhaps looking at both electrical schematics of both is a good alternative

rewolff commented 6 years ago

I'm slowly starting to look into sources of klipper. It seems that the "total shutdown" that I dislike from Marlin is NOT present in klipper. I've found the code to maintain certain commands even when in "error shutdown". Good stuff.

C4g2 commented 6 years ago

Apologies for newbing all over the place, but I thought this was a thread about if the Anet A8 could run the Klipper firmware. Id be more than willing to experiment with my hardware if no ones gotten it working with this model, but Im afraid ill need a bit more instruction. This is my first 3d printer and I've only done enough research to know that Klipper is the firmware I want to run/learn.

In that light: What same ramp settings? a stock a8? a similar printer that does run klipper thats 'close enough'? And again, which 'both' schematics?

I very much appreciate the response none the less!

ps. Your original post, 'has support for a8' was that a statement or a question? I think that kicked off my confusion.

KevinOConnor commented 6 years ago

On Sun, Dec 31, 2017 at 11:52:21PM +0000, C4g2 wrote:

Apologies for newbing all over the place, but I thought this was a thread about if the Anet A8 could run the Klipper firmware. Id be more than willing to experiment with my hardware if no ones gotten it working with this model, but Im afraid ill need a bit more instruction. This is my first 3d printer and I've only done enough research to know that Klipper is the firmware I want to run/learn.

Earlier, you indicated the electronics used an ATMEGA1284p. This is a micro-controller that Klipper supports. You should be able to follow the instructions in the docs/Installation.md document to compile and flash your electronics.

Your next step is going to be putting together a config file for the printer. This requires knowing which pins are attached to which hardware features. (For example, the x axis step pin might be "ar54".) This is pretty straight forward if you've got access to a schematic for the board, or if you know of some other common firmware that has support for the board that you can use as a reference.

In that light: What same ramp settings? a stock a8? a similar printer that does run klipper thats 'close enough'? And again, which 'both' schematics?

RAMPS is a common type of board (config/generic-ramps.cfg). I would verify that your board uses the same pin mapping before trying to use it though.

-Kevin

rewolff commented 6 years ago

Atmel branded ATMEGA1284p AU1720

That's an atmega-128-4-p-AU . 128 for 128 kb of memory. 4 for the 44-pin variant. p for the newer version that probably runs from 1.8V up to 5V and up to 20Mhz and AU for "TQFP" package. Yes, that should be compatible. It was produced 2017, week 20. (may 14-may 20)

diekaines commented 6 years ago

Very much interested if this would have Anet A8 config/support as well . its a replica printer ( prusa mk2 i believe ) and a low price ( ~ 180 USD ) so its popular for entry-level 3D printer https://www.banggood.com/Anet-A8-3D-Printer-DIY-Kit-1_75mm-0_4mm-Support-ABS-PLA-HIPS-p-1130694.html

Here the link on that same URL above for the docs https://1drv.ms/u/s!AuL7Lfc4swRWgR9OJnM0kLCzUM9m

Thank you

rewolff commented 6 years ago

I've seen even cheaper. (but maybe I didn't look carefully enough that something essential/expensive is missing). It is "I3" like.

Yes, it would be nice if someone would create a config file for these. However, this is not easily done "over the internet". You need to be close to the device to reverse-engineer the pinout.

Kevin, for initial config it would be nice to have some sort of debugging mode. In that mode it would be nice to be able to toggle and pulse arbitrary inputs and outputs to try and find those that control things. As I understand things now, the protocol between the server and the hardware already allows this. Not sure if the /tmp/printer protocol allows it. Then a separate application (web based?) can be used to show all inputs and allow control over all outputs. Click on output "D8" and suddenly the analog value on some analog pin (I don't know which by heart) starts to change and... the bed starts heating up. That sort of stuff. Of course, without feedback this is a bit dangerous, so proper warnings should be issued. ....

C4g2 commented 6 years ago

Thanks for the detailed reply! Knowing that it Will work with the A8 chip is great to hear, very encouraging!

News when Im stuck or have it.

manotroll commented 6 years ago

in marlin has the ramps settings files and A8 can be a good place to base

lenne0815 commented 6 years ago

Reowulf, theres a working A8 pinout for marlin, pins can just be copied from there.

KevinOConnor commented 6 years ago

On Mon, Jan 01, 2018 at 12:50:27PM +0000, Roger Wolff wrote:

Kevin, for initial config it would be nice to have some sort of debugging mode. In that mode it would be nice to be able to toggle and pulse arbitrary inputs and outputs to try and find those that control things. As I understand things now, the protocol between the server and the hardware already allows this. Not sure if the /tmp/printer protocol allows it. Then a separate application (web based?) can be used to show all inputs and allow control over all outputs. Click on output "D8" and suddenly the analog value on some analog pin (I don't know which by heart) starts to change and... the bed starts heating up. That sort of stuff. Of course, without feedback this is a bit dangerous, so proper warnings should be issued. ....

There is a console tool that can be used to issue low-level commands (see docs/Debugging.md). I wouldn't recommend trying to find pins by trial and error though.

As lenne0815 stated, there appears to be a Marlin config for an anet board, so I'd start with that.

-Kevin

diekaines commented 6 years ago

Dont know how relevant this might be. Output of Octoprint when connected to printer 

2018-01-02 18:06:16,896 - octoprint.util.comm - INFO - Printer reports firmware name "ANET_A8_20160525  http://www.anet3d.com/"
2018-01-02 18:06:16,898 - octoprint.util.comm - INFO - Detected Repetier firmware, enabling relevant features for issue free communication
lenne0815 commented 6 years ago

I think to make it work you need to let klipper generate the firmware hex and flash these via atmel studio to the board, maybe someone can confirm this ?

dragonnn commented 6 years ago

First of - learn how to flash you board for example by flashing Marlin with is official supported. then try to learn how clipper works.

tilsonm commented 6 years ago

I think to make it work you need to let klipper generate the firmware hex and flash these via atmel studio to the board, maybe someone can confirm this ?

@lenne0815 I managed to compile the hex, pull it to my Windows desktop, and flash the board using atmel studio, but it soft-bricked my board. I was able to re-flash Marlin 1.1.8, though. Tried with avrdude, and same result.

Here is the output from avrdude.

avrdude: Version 6.3, compiled on Jan 17 2017 at 12:00:53
         Copyright (c) 2000-2005 Brian Dean, http://www.bdmicro.com/
         Copyright (c) 2007-2014 Joerg Wunsch

         System wide configuration file is "C:\Program Files (x86)\Arduino\hardware\tools\avr/etc/avrdude.conf"

         Using Port                    : COM3
         Using Programmer              : arduino
         Overriding Baud Rate          : 57600
         AVR Part                      : ATmega1284P
         Chip Erase delay              : 55000 us
         PAGEL                         : PD7
         BS2                           : PA0
         RESET disposition             : dedicated
         RETRY pulse                   : SCK
         serial program mode           : yes
         parallel program mode         : yes
         Timeout                       : 200
         StabDelay                     : 100
         CmdexeDelay                   : 25
         SyncLoops                     : 32
         ByteDelay                     : 0
         PollIndex                     : 3
         PollValue                     : 0x53
         Memory Detail                 :

                                  Block Poll               Page                       Polled
           Memory Type Mode Delay Size  Indx Paged  Size   Size #Pages MinW  MaxW   ReadBack
           ----------- ---- ----- ----- ---- ------ ------ ---- ------ ----- ----- ---------
           eeprom        65    10   128    0 no       4096    8      0  9000  9000 0xff 0xff
           flash         65    10   256    0 yes    131072  256    512  4500  4500 0xff 0xff
           lock           0     0     0    0 no          1    0      0  9000  9000 0x00 0x00
           lfuse          0     0     0    0 no          1    0      0  9000  9000 0x00 0x00
           hfuse          0     0     0    0 no          1    0      0  9000  9000 0x00 0x00
           efuse          0     0     0    0 no          1    0      0  9000  9000 0x00 0x00
           signature      0     0     0    0 no          3    0      0     0     0 0x00 0x00
           calibration    0     0     0    0 no          1    0      0     0     0 0x00 0x00

         Programmer Type : Arduino
         Description     : Arduino
         Hardware Version: 2
         Firmware Version: 1.16
         Vtarget         : 0.0 V
         Varef           : 0.0 V
         Oscillator      : Off
         SCK period      : 0.1 us

avrdude: AVR device initialized and ready to accept instructions

Reading | ################################################## | 100% 0.00s

avrdude: Device signature = 0x1e9705 (probably m1284p)
avrdude: safemode: hfuse reads as 0
avrdude: safemode: efuse reads as 0
avrdude: reading input file "C:\Users\tilso\Desktop\klipper.elf.hex"
avrdude: writing flash (13338 bytes):

Writing | ################################################## | 100% 3.78s

avrdude: 13338 bytes of flash written
avrdude: verifying flash memory against C:\Users\tilso\Desktop\klipper.elf.hex:
avrdude: load data flash data from input file C:\Users\tilso\Desktop\klipper.elf.hex:
avrdude: input file C:\Users\tilso\Desktop\klipper.elf.hex contains 13338 bytes
avrdude: reading on-chip flash data:

Reading |                                                    | 0% 0.00savrdude: stk500_loadaddr(): (a) protocol error, expect=0x14, resp=0x49

avrdude: stk500_paged_load(): (a) protocol error, expect=0x14, resp=0x4e
Reading | #                                                  | 1% 0.05savrdude: stk500_cmd(): programmer is out of sync
avr_read(): error reading address 0x0000
    read operation not supported for memory "flash"
avrdude: failed to read all of flash memory, rc=-2

avrdude: stk500_cmd(): programmer is out of sync
avrdude: stk500_cmd(): programmer is out of sync
avrdude: stk500_cmd(): programmer is out of sync
avrdude: stk500_cmd(): programmer is out of sync
avrdude: safemode: Sorry, reading back fuses was unreliable. I have given up and exited programming mode
avrdude: stk500_disable(): protocol error, expect=0x14, resp=0x2e

avrdude done.  Thank you.
tilsonm commented 6 years ago

@lenne0815 I've successfully flashed my Anet v1.0 with klipper. I have no idea what changed, but after flashing back to Marlin, I tried avrdude again and it flashed flawlessly. I think that it worked the first time, but I mistakenly thought that it not being able to read back the flash data and then the white squares on the display that it was soft-bricked.

Testing now.

C4g2 commented 6 years ago

@Tilsonm Thats great to hear, what config file did you use when building the hex? Best of luck on the test!

@Lenne0815 Hey, just recognized your name from Hackaday comments on this firmware, thanks for the videos you posted!

tilsonm commented 6 years ago

@C4g2 I ran menuconfig with Atmega AVR, atmega1284p, 16MHz, 115200 baud rate. When I flashed the board, I had to modify the flashing baudrate to 57600, and it has to be flashed manually with avrdude or with Atmel Studio, or at least that's the only way I could get it to flash.

tilsonm commented 6 years ago

Good news, all. I've got klipper working on the stock Anet v1 board and have verified the board configuration. I'm going to do some testing to see if I can work out some "optimal" settings, but it works. I will try to write up a quick guide in the next few days so others can try to replicate it on their Anet v1 boards.

Quick question: any hopes of the LCD screen working? Has a generic framework been implemented yet?

C4g2 commented 6 years ago

Across my research I think I've found mention of atleast 3 different layouts, yet still called Anet V1 boards, with mainly just the addition of trim pots for some or all of the stepper drivers. The specific board I have has a single trim pot for the Z stepper driver, if anyone has suggestions how to differentiate between them when talking, please help. For now I've been calling mine what is printed on the board, even tho it makes no sense, the "Anet V1-5." One through Five? One dash Five? whatever...

Point is I found a schematic https://github.com/ralf-e/ANET-3D-Board-V1.0/blob/master/ANET3D_Board_Schematic.png and have put together what i believe is a correct config file based on this schematic and a few online calculators for step per mm and whatnot. I'll be reporting back with how the first run goes as soon as I have it up.

@tilsonm I thought the 1284p was a 20mhz chip? Thats too bad to hear about the RaspPi flash not working tho.

rewolff commented 6 years ago

I thought the 1284p was a 20mhz chip? Thats too bad to hear about the RaspPi flash not working tho.

From the "world of arduino" there is some legacy that even though the chips can do 20MHz,. they are clocked at only 16Mhz. For arduino the origin is that they started with an atmega8 chip, which is only 16Mhz. The upgrade to atmega168 allowed 20Mhz, but they decided to keep as much software compatibility by keeping the 16Mhz clock frequency.

For the atmega1286 and the likes, the chip might be able to do other frequencies, but to do USB they might need a specific crystal. Chips like STM32F072 allow a 4-24Mhz crystal (or none!) and then still they can do USB. The ATMEL chips require one of a few specific crystals: 8 or 16 to be able to do USB. If you decide 12MHz (which is what the FT232 requires to be able to do USB) then the atmel simply can't do USB. Simple as that.

I might be talking about a slighly older version of the atmel chips and that they've updated their USB module. in this chip allowing different crystals. But in any case, the question in the printer.cfg file is not: "what can your chip do?" but "what crystal is installed on your board" ?

@c4g2: Try selecting all the text, and hitting "code" (the angled brackets) above the edit window.

C4g2 commented 6 years ago

Thank you for the clarification, that could have been a small headache to figure out! My last post just doesnt wanna work for me. I tried a few things, including the code button, before giving up and blaming it on my mac running an old version of chrome. But thank you for trying.

I'll try flashing with the rasppi just to make sure it wasnt a problem with tilsonm's specific hardware, if it still fails, avrdude then load this mess of a config file and give it a go!

C4g2 commented 6 years ago

One last try at config file:

Anet_A8.cfg

# This file is for the Anet A8 boards, based on the Anet V1-5 board.
# All versions come populated with the Atmel ATMEGA1284P.
#
# Pin names may be preceded by an '!' to indicate that a reverse
# polarity should be used (eg, trigger on low instead of high). Input
# pins may be preceded by a '^' to indicate that a hardware pull-up
# resistor should be enabled for the pin.

# The stepper_x section is used to describe the stepper controlling
# the X axis in a cartesian robot.
[stepper_x]
step_pin: PD7
dir_pin: PC5
enable_pin: !PD6
step_distance: 0.15385
endstop_pin: ^PC2
#position_min: 0
position_endstop: 0
position_max: 220
homing_speed: 25
#homing_retract_dist: 5.0
#homing_positive_dir: False
#homing_stepper_phases: 0
#homing_endstop_accuracy: 0.200
#homing_endstop_phase: 0
#homing_endstop_align_zero: False

# The stepper_z section is used to describe the stepper controlling
# the Z axis in a cartesian robot.
[stepper_y]
step_pin: PC7
dir_pin: PC6
enable_pin: !PD6
step_distance: 0.15385
endstop_pin: ^PC3
#position_min: 0
position_endstop: 0
position_max: 220
homing_speed: 25
#homing_retract_dist: 5.0
#homing_positive_dir: False
#homing_stepper_phases: 0
#homing_endstop_accuracy: 0.200
#homing_endstop_phase: 0
#homing_endstop_align_zero: False

# The stepper_z section is used to describe the stepper controlling
# the Z axis in a cartesian robot.
[stepper_z]
step_pin: PB4
dir_pin: PB2
enable_pin: !PA5
step_distance: 0.004
endstop_pin: ^PC4
#position_min: 0
position_endstop: 0.25
position_max: 240
#homing_speed: 5.0
#homing_retract_dist: 5.0
#homing_positive_dir: False
#homing_stepper_phases: 0
#homing_endstop_accuracy: 0.200
#homing_endstop_phase: 0
#homing_endstop_align_zero: False

# The extruder section is used to describe both the stepper
# controlling the printer extruder and the heater parameters for the
# nozzle.
[extruder]
step_pin: PB1
dir_pin: PB0
enable_pin: !PD6
step_distance: .002
# The remaining variables describe the extruder heater.
nozzle_diameter: 0.400
filament_diameter: 1.750
#max_extrude_cross_section: 0.64
#max_extrude_only_distance: 50.0
#max_extrude_only_velocity:
#max_extrude_only_accel:
#pressure_advance: 0.0
#pressure_advance_lookahead_time: 0.010
heater_pin: !PD5
#max_power: 1.0
sensor_type: NTC 100K beta 3950
sensor_pin: ^PA7
#pullup_resistor: 4700
#adc_voltage: 5.0
control: pid
pid_Kp: 22.2
pid_Ki: 1.08
pid_Kd: 114
#pid_deriv_time: 2.0
#pid_integral_max: 1.0
#min_extrude_temp: 170
min_temp: 0
max_temp: 250

# The heater_bed section describes the heated bed.
[heater_bed]
heater_pin: !PD4
sensor_type: NTC 100K beta 3950
sensor_pin: ^PA6
control: watermark
max_delta: 2.0
min_temp: 0
max_temp: 100

#Front Mounted Part Cooling Fan
[fan]
pin: PB4
#max_power: 1.0
#hard_pwm: 0
#kick_start_time: 0.100

#Side Mounted Extruder Fan
[heater_fan extruder_fan]
pin: PD0
heater: extruder
#heater_temp: 50.0
#fan_speed: 1.0
#max_power: 1.0
#hard_pwm: 0
#kick_start_time: 0.100

# Micro-controller information.
[mcu]
serial: /dev/ttyACM0
baud: 115200
#pin_map: -Removed 'arduino'
#restart_method: arduino

# The printer section controls high level printer settings.
[printer]
kinematics: cartesian
max_velocity: 300
max_accel: 3000
#max_accel_to_decel: 1500
max_z_velocity: 5
max_z_accel: 100
#motor_off_time: 600
#junction_deviation: 0.02
manotroll commented 6 years ago

how many people testing and I still do not have my printer to help with the tests: \

how to solve the recording problem since the board runs at 20mhz had to change something in arduino IDE?

lenne0815 commented 6 years ago

@tilsonm Theres neither need nor plans to implement the lcd displays afaik. Look at Printoid for a great gui instead.

tilsonm commented 6 years ago

@C4g2 I suggest doing an ls -l /dev/serial/by-id and getting the exact path to your printer. I turn my printer off but leave my Pi on, and when I restart the symlinked /dev/tty path has changed. Oddly, mine is /dev/ttyUSB, not /dev/ttyACM*

Also, the A8 has a head offset, so that will need to be taken care of in your cfg to get correct positioning.

See section from FAQ for setting the address more concretely https://github.com/KevinOConnor/klipper/blob/master/docs/FAQ.md

EDIT: Here is my printer.cfg for the Anet v1 and have confirmed everything working except my extruder is wonky and needs some work.

`# This file contains common pin mappings for Melzi v2.0 boards. To use
# this config, the firmware should be compiled for the AVR
# atmega1284p.

# Note that the "make flash" command does not work with Melzi
# boards. The boards are typically flashed with this command:
#  avrdude -p atmega1284p -c avrisp -P /dev/ttyUSB0 -U out/klipper.elf.hex
# or on some boards with:
#  avrdude -p atmega1284p -c arduino -P /dev/ttyUSB0 -U out/klipper.elf.hex

# See the example.cfg file for a description of available parameters.

[stepper_x]
step_pin: PD7
dir_pin: PC5
enable_pin: !PD6
step_distance: .01
endstop_pin: ^!PC2
position_endstop: -20
position_max: 220
position_min: -20
homing_speed:  100

[stepper_y]
step_pin: PC6
dir_pin: PC7
enable_pin: !PD6
step_distance: .01
endstop_pin: ^!PC3
position_endstop: -8
position_min: -8
position_max: 208
homing_speed: 100

[stepper_z]
step_pin: PB3
dir_pin: !PB2
enable_pin: !PA5
step_distance: 0.0025
endstop_pin: ^!PC4
position_endstop: 0.5
position_max: 200
homing_speed: 20

[extruder]
step_pin: PB1
dir_pin: PB0
enable_pin: !PD6
step_distance: .002
nozzle_diameter: 0.400
filament_diameter: 1.750
heater_pin: PD5
sensor_type: ATC Semitec 104GT-2 
sensor_pin: PA7
control: pid
pid_Kp: 22.2
pid_Ki: 1.08
pid_Kd: 114
min_temp: 0
max_temp: 250

[heater_bed]
heater_pin: PD4
sensor_type: ATC Semitec 104GT-2
sensor_pin: PA6
control: watermark
min_temp: 0
max_temp: 130

[fan]
pin: PB4

[mcu]
serial: /dev/serial/by-id/usb-1a86_USB2.0-Serial-if00-port0
baud: 57600 

[printer]
kinematics: cartesian
max_velocity: 300
max_accel: 3000
max_z_velocity: 5
max_z_accel: 100
C4g2 commented 6 years ago

Will do, mine is /ttyUSB0 as well, and thank you for posting your config file! Copied most your axis measurements & offsets (Z isn't ~240?), and also helped me notice a typo at my z step pin.

About the sensor type; I've not found much info on what comes stock in this MK8 extruder, I did find more than one place mentioning the NTC 3950, but that could of been meant as an upgrade from stock equipment. If Im reading the schematic correctly, both of the sensors have 4.7k pull ups on them. But you confirmed End and Bed temp both reporting correctly?

Anyone else lurking this thread now or later, if you're looking to program your printers board and dont have a programmer, apparently you can [program your AVR via RaspPi's GPIO] (https://learn.adafruit.com/program-an-avr-or-arduino-using-raspberry-pi-gpio-pins/overview) have yet to try it, but it looks solid.

C4g2 commented 6 years ago

I can also confirm that make flash does not work via rasppi. Tried both /dev/ttyUSB0 and the /by-id/**** at 250k 115k and 57k, regardless, printer restarts then times out.

Is this a Bootloader issue, or something else, hopefully more easily fixed?

rewolff commented 6 years ago

"make flash" tries to program the MCU in the printer. Normally, a program like "AVRDUDE" talks to "programming hardware", which then programs the target CPU. The original thinking in "arduino" was that an arduino would be simple because you would not need a separate programmer. This was accomplished by burning a program into the arduino that would PRETEND to be a programmer, while infact programming the CPU itself. So ESP8266 and STM32 have a boot loader protocol that is different from what hardware progammer modules use, but here the bootloader simply emulates being a programmer.

On my system this worked: my MCU was tricked to boot into the bootloader, and my pi talked to the MCU in the "programmer-protocol" while the bootloader pretended to be programming a different MCU while in fact it was self-programming.

This works because my board is actually an arduino with a bootloader. But if you are selling hundreds of A8 printers, you get to chose: "Do we program the bootloader AND the default firmware, or do we just program the firmware?"

You can bet your ass that the choice was made to program just the firmware.

So when you issue "make flash" the makefile will try to program your MCU by trying to get it into the bootloader (not present, so it will immediately boot back into the default firmware) and then try to talk to it using the bootloader/programmer protocol. That'll time out.

C4g2 commented 6 years ago

So first thank you for the answer! But that leads me to the question is it worth putting a 'standard' bootloader on there? Is there even such a thing? Cause if klippers firmware portion is gonna be undergoing many revisions, it sounds like it might be worth the time to get the rasppi flashing it properly. I know this is almost outside the scope of klipper, but im curious : )

KevinOConnor commented 6 years ago

On Thu, Jan 04, 2018 at 02:40:23AM -0800, C4g2 wrote:

So first thank you for the answer! But that leads me to the question is it worth putting a 'standard' bootloader on there? Is there even such a thing? Cause if klippers firmware portion is gonna be undergoing many revisions, it sounds like it might be worth the time to get the rasppi flashing it properly. I know this is almost outside the scope of klipper, but im curious : )

Yes. If your board did not come with a bootloader installed, then you will need to use an external programmer to program the board. In that process, you should write a bootloader to the board (there is no downside to having a bootloader). Then it should be possible to program future firmware updates without requiring an external programmer.

-Kevin

tilsonm commented 6 years ago

@C4g2, lenne0815 reminded me today that I never shared my "secret formula" of avrdude commands to flash the hex file to my Anet board.

Here goes (written as if I was doing this again for the first time):

  1. ssh into pi using PuTTY.
  2. git clone https://github.com/KevinOConnor/klipper
  3. ./klipper/scripts/install-octopi.sh
  4. cd ~/klipper/
  5. make menuconfig
  6. Select options: MCU = Atmega AVR; Processor Model = atmega1284p; Processor speed = 16MHz; Baud Rate = 115200. Save using default file name suggested by utility. Exit utility.
  7. make
  8. sudo service klipper stop

    Here's where the REAL magic starts

  9. avrdude -v -p atmega1284p -c arduino -P /dev/ttyUSB# -b 57600 -D -U flash:w:/home/pi/klipper/out/klipper.elf.hex:i

Explanation of arguments (to the best of my ability):

Then you can run sudo service klipper start and follow the rest of Kevin's instructions in the Installation page.

C4g2 commented 6 years ago

@Tilsonm Awesome guide, the flag explanations help with whats in man avrdude. I would advise anyone like me new to flashing chips that the Anet board is not particularly friendly with its programmer header; upside down and only the middle 6, good pics and other sage advice about this printer there too.

@Kevin Programming the board doesnt seem to be any problem, but might as well do it "hard" once and easy the rest. Optiboot was the first loader I found, hes looking for folk with the 1284p to test it, but I have no idea what to look for in terms of requirments or compatibility, essentially if Optiboot is even an option for this project.

Other news, I got the link I posted previously about AVR programming with RaspPi GPIO pins working, and just had my first success at a -v and getting a response from the chip with a 40-pin ribbon cable and 6 dupont male to female wires. Figured I'd wait to find something like Tilsonm's guide, thanks again, before I finally flashed anything. Now I think I'll try a bootloader option first and see if the rasppi can do this via usb.

Any suggestions?

hg42 commented 6 years ago

you can try any boot loader, you can always flash another.

You eventually need to program the fuses, because different boot loaders often start at different addresses. https://embedderslife.wordpress.com/2012/08/20/fuse-bits-arent-that-scary/ http://www.engbedded.com/fusecalc/

You can fake to flash a boot loader with the arduino ide. This will output command lines in the terminal that you can use as hint, which values you need. You have to choose the correct board and cpu. Eventually you need to install additional board managers for this.

These command lines are for the Arduino standard boot loader, so you may have to change the fuses for the start address. You can see which meaning the values have and also change options and calculate new values with the fusecalc in the second link.

With my TronXY board I had best success with the boot loaders in the arduino ide (see command lines to find them).

rewolff commented 6 years ago

-c means the programmer used. In this case it's arduino. All the research I can find says it should be avrispmkII, but I got the terminal command from ArduinoIDE on a successful flash of Marlin and even though I selected avrispmkII in ArduinoIDE, the command programmed with arduino.

I explained that the bootloader in the arduino pretends to be a programmer. Well, my memory for such details is bad, but that programmer is apparently 'avrispmkII'. When you specify "arduino" as the programmer, then avrdude will follow the AVRISPMKII protocol, but before starting the protocol avrdude will issue a few tricks that will cause an arduino to restart-into-bootloader. If your anet responds to this with a succesfull flash, you've proven me wrong in that there IS a bootloader in your board. :-)

In my case it's /dev/ttyUSB0 or /dev/ttyUSB1 (it changes each time I power up my printer).

The Linux kernel will chose the first free /dev/ttyUSBxx with xx being a number starting at 0. Now if you unplug the printer while it is connected as "ttyUSB0", it is likely that SOME program is still holding a filedescriptor to that device. So at that moment, the /dev/ttyUSB0 device file will be removed, but the index "0" is not yet marked as free. If you replug the device, it will get index 1: /dev/ttyUSB1. At that moment you notice something isn't working and you reconfigure klipper to use /dev/ttyUSB1 and it releases the filedescriptor and the index becomes free again.

rewolff commented 6 years ago

(there is no downside to having a bootloader).
There is very little downside to having a bootloader.... The small downside is that you lose a few bytes of your flash memory to the bootloader. With the firmware fitting easily in what's left, for klipper there is indeed no downside.... :-)

tilsonm commented 6 years ago

@rewolff that makes sense. What you said the first time didn't quite click, but thanks for repeating it. It makes more sense now.

How can I fix the ttyUSB#. I thought the suggestion I made to c4g2 about ls -l /dev/serial/by-id would take care of it, but it still seems confused when I power my printer off and back on and switches between USB0 and USB1. I've tried symlinking to /dev/anetv1 by using USB characteristics, but for some reason it's still not recognizing and automatically adding that symlink. Or maybe I'm going about this all wrong.

KevinOConnor commented 6 years ago

On Fri, Jan 05, 2018 at 12:41:14PM +0000, tilsonm wrote:

How can I fix the ttyUSB#. I thought the suggestion I made to c4g2 about ls -l /dev/serial/by-id would take care of it, but it still seems confused when I power my printer off and back on and switches between USB0 and USB1. I've tried symlinking to /dev/anetv1 by using USB characteristics, but for some reason it's still not recognizing and automatically adding that symlink. Or maybe I'm going about this all wrong.

Did you follow the directions at: https://github.com/KevinOConnor/klipper/blob/master/docs/FAQ.md#wheres-my-serial-port

If so, can you provide the klippy.log file with the failure (along with ls -l /dev/serial/by-id/ after the failure).

-Kevin

KevinOConnor commented 6 years ago

On Thu, Jan 04, 2018 at 04:22:05PM -0800, tilsonm wrote:

@C4g2, lenne0815 reminded me today that I never shared my "secret formula" of avrdude commands to flash the hex file to my Anet board.

Here goes (written as if I was doing this again for the first time): [...]

Thanks. I've updated the Klipper build so that it will always use "-c arduino" for the atmega1284p (commit 5a86391f). So, once one gets a bootloader on the board, it should be possible to use the standard "make flash" instructions to flash Klipper.

I don't have the board, but this link seems helpful for getting a bootloader installed: http://www.instructables.com/id/Flashing-a-Bootloader-to-the-CR-10/

-Kevin

C4g2 commented 6 years ago

@kevin Thanks, good link too, I found the bootloader the talking about and judging by the file name, it looks to support the 1284p!

So the .hex is what I flash with avrdude using -U boot:w:~/optiboot_atmega1284p.hex:i? There is also a .lst file in the parent folder of the .hex that has the exact same name, do I need to do anything with that?

C4g2 commented 6 years ago

I was thinking about Tilsonm's last post, I'm not at home right now so I cant try this myself, but I was thinking about symlinking the /dev/serial/by-id to /dev/anet? udev should have a rules file you could try hardcoding it in if you're brave!

tilsonm commented 6 years ago

@c4g2 I tried using udev and couldn't get it to work. Lol. You might have better luck.

rewolff commented 6 years ago

There is also a .lst file in the parent folder of the .hex that has the exact same name, do I need to do anything with that?

No. The .lst will list where things have ended up in memory. i.e. if you declare a variable "int steps_to_make", the LST will describe what address in memory is now used for storing that variable.