Getting this to work

[siredmar]

Hi,

i'm triying to get HAL2Arduino to work and i'm stuck at some point here.

What i have done is download HAL2Arduino-0.5 and change the pinning and stepsPerInchX/Y/Z for my machine (modAccelStepper ) and the limit switches (modLimSwitch). I compiled it and flashed it to my arduino nano (with Atmega 328p). I connected my cnc to the configured pins. Then i went to linuxcnc and get the Ubuntu 10.04 package with all linuxcnc related stuff included and installed it in a virtualbox.
I followed the instrucions.txt from your github repo and created a new config 'my-mill', changed the path for the python script and modified the config that linuxcnc will find the script. The Arduino is connected to the virtualbox and it's been recgnized with /dev/ttyUSB0.

When i try to start linuxcnc with the shortcut on the desktop for my configuration i receive the following errors:
Can not find -sec MOT -var MOD -num 1
Can not find -sec IO - var IO -num 1
Can not find -sec LINUXCNC -var NML_FILE -num 1
Can not find -sec EMC -var NML_FILE -num 1
Waiting for component 'Hal2Arduino' to become ready......................................Traceback )most recent call last):
File "/opt/HAL2Arduino-0.5.py", line 2020, in
ser.close()
NameError: global name 'ser' is not defined

custom.hal:2. /opt/HAL2Arduino-0.5.py exited without becoming ready
1474
PID TTY STAT TIME COMMAND
Stopping realtime threads
Unloading hal components

Can anyone tell me what's going on there?

[siredmar]

I got one step further. I modified the python script to set the device "/dev/ttyUSB0" as port... But the message is still there:
Waiting for component 'Hal2Arduino' to become ready......................................Traceback )most recent call last):

But one thing seems good: the red uart led flashes on startup of linuxcnc.

[rene-dev]

On 02 Jun 2014, at 21:42, siredmar notifications@github.com wrote:

I got one step further. I modified the python script to set the device "/dev/ttyUSB0" as port... But the message is still there:
Waiting for component 'Hal2Arduino' to become ready......................................Traceback )most recent call last):

I had to modify this line as well.
This is how far I get:

halcmd: loadusr -Wn ard /home/emc/HAL2Arduino-0.5.py
Waiting for component 'ard' to become ready............Found device on /dev/ttyUSB0
Asking for firmware info.
Firmware: HAL-2-Arduino
Unit: '0'
Version: 0.5
.comThread: port=/dev/ttyUSB0 arrayPos:0 Queue:0x8d22b2c
.....................................makePins: creating: 'axis_0_home-sw-in'
makePins: creating: 'axis_0_joint-pos-cmd'
makePins: creating: 'axis_0_pos-lim-sw-in'
makePins: creating: 'axis_0_neg-lim-sw-in'
makePins: creating: 'axis_1_home-sw-in'
makePins: creating: 'axis_1_joint-pos-cmd'
makePins: creating: 'axis_1_pos-lim-sw-in'
makePins: creating: 'axis_1_neg-lim-sw-in'
makePins: creating: 'axis_2_home-sw-in'
makePins: creating: 'axis_2_joint-pos-cmd'
makePins: creating: 'axis_2_pos-lim-sw-in'
makePins: creating: 'axis_2_neg-lim-sw-in'
commandList=['']
linkData=''
listOfPorts=['/dev/ttyUSB0']
listOfCommands=['215 224 231 233']
listOfAxis=['0 1 2']
listOfUnits=['0']
linkMsg=''
comService: started.
comThread: Sending 996 0 0 996; to /dev/ttyUSB0
………………………………………………..

hangs here, printing dots

Rene

[dewy721]

Looks like to .py file is not returning the .ready back to hal to signal
that its up and running.
On Jun 3, 2014 4:43 AM, "Rene Hopf" notifications@github.com wrote:

On 02 Jun 2014, at 21:42, siredmar notifications@github.com wrote:

I got one step further. I modified the python script to set the device
"/dev/ttyUSB0" as port... But the message is still there:
Waiting for component 'Hal2Arduino' to become
ready......................................Traceback )most recent call
last):

I had to modify this line as well.
This is how far I get:

halcmd: loadusr -Wn ard /home/emc/HAL2Arduino-0.5.py
Waiting for component 'ard' to become ready............Found device on
/dev/ttyUSB0
Asking for firmware info.
Firmware: HAL-2-Arduino
Unit: '0'
Version: 0.5
.comThread: port=/dev/ttyUSB0 arrayPos:0 Queue:0x8d22b2c
.....................................makePins: creating:
'axis_0_home-sw-in'
makePins: creating: 'axis_0_joint-pos-cmd'
makePins: creating: 'axis_0_pos-lim-sw-in'
makePins: creating: 'axis_0_neg-lim-sw-in'
makePins: creating: 'axis_1_home-sw-in'
makePins: creating: 'axis_1_joint-pos-cmd'
makePins: creating: 'axis_1_pos-lim-sw-in'
makePins: creating: 'axis_1_neg-lim-sw-in'
makePins: creating: 'axis_2_home-sw-in'
makePins: creating: 'axis_2_joint-pos-cmd'
makePins: creating: 'axis_2_pos-lim-sw-in'
makePins: creating: 'axis_2_neg-lim-sw-in'
commandList=['']
linkData=''
listOfPorts=['/dev/ttyUSB0']
listOfCommands=['215 224 231 233']
listOfAxis=['0 1 2']
listOfUnits=['0']
linkMsg=''
comService: started.
comThread: Sending 996 0 0 996; to /dev/ttyUSB0
………………………………………………..

hangs here, printing dots

Rene

—
Reply to this email directly or view it on GitHub
#3 (comment).

[siredmar]

I got one step further (by accident)... When i flash the HAL2Arduino hexfile and try directly after that, the python script detects the hal2arduino firmware. After that the firmware version should be sent, but never returns. It seems like the pairing will not work as planned... btw i use the arduino nano with just moAccelStepper and modLimitSwitch mods...

[dewy721]

Ah. I see. The python file expects the arduino to respond in a preset
amount of time. If its just a tiny bit too slow it won't connect.
On Jun 3, 2014 3:49 PM, "siredmar" notifications@github.com wrote:

I got one step further (by accident)... When i flash the HAL2Arduino
hexfile and try directly after that, the python script detects the
hal2arduino firmware. After that the firmware version should be sent, but
never returns. It seems like the pairing will not work as planned... btw i
use the arduino nano with just moAccelStepper and modLimitSwitch mods...

—
Reply to this email directly or view it on GitHub
#3 (comment).

[siredmar]

So can you give me a hint where to tune the timings? Should i modify it in the python script?
Or can be a proper handshake mechanism implemented in the arduino/python code?

[dewy721]

Near line#121 of the .py file....
Change the serial packet timout value to timeout=0.05 or higher. Note: the
higher the value the easier to connect but also more latency of the serial
link.
On Jun 3, 2014 4:01 PM, "siredmar" notifications@github.com wrote:

So can you give me a hint where to tune the timings? Should i modify it in
the python script?
Or can be a proper handshake mechanism implemented in the arduino/python
code?

—
Reply to this email directly or view it on GitHub
#3 (comment).

[siredmar]

I tried timeout=0.06 and 0.5 and 1 and 2 but no effect. The only thing that is working is after plugging in the usb cable directly start linuxcnc. It seems to search the firmware, but nothing is received...
Starting LinuxCNC...
Waiting for component 'Hal2Arduino' to become ready...........Found device on /dev/ttyUSB0
Asking for firmware info.
Firmware: Failed. HAL-2-Arduino not found on this device
...............................................................Wait, something is not right here.

But if i try again without reconnecting the arduino, it won't work again

[dewy721]

It will also do that if the computer has another arduino that is not loaded
with hal2arduino. I really need to fix that. :-/
On Jun 3, 2014 4:19 PM, "siredmar" notifications@github.com wrote:

I tried timeout=0.06 and 0.5 and 1 and 2 but no effect. The only thing
that is working is after plugging in the usb cable directly start linuxcnc.
It seems to find the arduino firmware, but the firmware version is not
received:
Starting LinuxCNC...
Waiting for component 'Hal2Arduino' to become ready...........Found device
on /dev/ttyUSB0
Asking for firmware info.
Firmware: Failed. HAL-2-Arduino not found on this device
...............................................................Wait,
something is not right here.

—
Reply to this email directly or view it on GitHub
#3 (comment).

[siredmar]

Btw: there is a fork of your work (https://github.com/CCHS-Melbourne/HackCNC) that hast the same goal as you... i tried that one and could connect to it pretty well and without problems... the code seems pretty stripped down from yours... don't know if thats a bug or a feature :)
Still i'd like to see hal2arduino running because imho its the better approach.

[dewy721]

I remember them, nice bunch. School kids actually. Its not so much as
stripped down. Just older, less developed version of my work. I'm guessing
that it's probably over a year ago that I helped them, smart group too.
On Jun 3, 2014 4:31 PM, "siredmar" notifications@github.com wrote:

Btw: there is a fork of your work (
https://github.com/CCHS-Melbourne/HackCNC) that hast the same goal as
you... i tried that one and could connect to it pretty well and without
problems... the code seems pretty stripped down from yours... don't know if
thats a bug or a feature :)
Still i'd like to see hal2arduino running because imho its the better
approach.

—
Reply to this email directly or view it on GitHub
#3 (comment).

[siredmar]

Did you ever experience such problems like i do connecting the hal2arduino?

[dewy721]

Since I wrote the code from scratch, yes. Many times.
On Jun 3, 2014 4:44 PM, "siredmar" notifications@github.com wrote:

Did you ever experience such problems like i do connecting the
hal2arduino?

—
Reply to this email directly or view it on GitHub
#3 (comment).

[siredmar]

Hi,

i debugged the code a little bit. The 'ok' message seems to be received. What is not working is the comSetup function as it won't pass the line ser.write(lastCommand). I'm no python expert, but i think it is somehow connected to pyserial... I use python 2.7.5 with pyserial 2.7.
Can you give me info on your setup?

[siredmar]

I stripped the python program down to the minimum and added some wait loop for handshake and i get the expected results:

#!/usr/bin/python
import sys, string
import serial
import time

firmware = 'HAL-2-Arduino'
firmwareVersion = 0.5
port = '/dev/ttyUSB0'

def comSetup(ser, command, data1, data2):
    lastChecksum=command + data1 + data2
    lastCommand="%s %s %s %s%s" % (str(command), str(data1), str(data2), str(lastChecksum), ";")
    ser.flush()
    ser.write(lastCommand)
    for i in range(0,6): # attemp to send a command, but no more then 5 times.
        linkMsg = ser.readline()
        if linkMsg.find('resend') > -1:
            print "resend"
            ser.flushInput()
            ser.write(lastCommand)
        else: # Command sent successfully.
            returnvalue = linkMsg.strip('\r\n')
            return returnvalue

ser = serial.Serial(port, 115200, timeout=1)
print "Found device on %s\nAsking for firmware info." % port
time.sleep(2)
linkMsg = 'notok'

while linkMsg.find('ok') != -1:
   linkMsg = ser.readline()
   time.sleep(0.5)

if comSetup(ser, 990, 0, 0) == firmware:
   print "firmware %s" % firmware
   unitId=comSetup(ser, 992, 0, 0)
   if unitId != "":
      print "Unit: %r" % unitId
      version=float(comSetup(ser, 991, 0, 0))
      print "version: %s" % version
      if version >= firmwareVersion:
         codesAccepted = comSetup(ser, 993, 0, 0)
         print "codesAccepted: %s" % codesAccepted
ser.close()

This is the output:
$ ./testserial.py
Found device on /dev/ttyUSB0
Asking for firmware info.
firmware HAL-2-Arduino
Unit: '0'
version: 0.5
codesAccepted:

So the communication seems to work with pyserial and my version of python. With this wait routine the arduino can be reset with the normal reset switch an the python script waits until the "ok" comes in. When i insert this exact same wait loop in the original py script, nothin happens. Maybe it's related to those threading functions it is using?