Advanced Search

Despite this Axis still fails to respond to the M61 Qn and reports a No tool status.

This is not a fix for M61, it is a modified iocontrol which allows the tool number to be updated via the 2 pins I described.

If you do not have a toolchanger component or some other method of triggering them, they will not be of use to you, sorry.

OK I now have understood and executed your instructions. I have replaced /usr/bin/io with
the io file you provided and just for good measure I performed a restarted of the computer.

Despite this Axis still fails to respond to the M61 Qn and reports a No tool status.
Have forgot some other detail???

Now I installed 2.6.0~pre to implement set-world-mode pin, and it seems that it now works... But it occurred that there is a problem with this wireless pendant compatibility with this version, I'm receiving this when trying to start up Linuxcnc (Run In Place):

xhc-hb04: symbol lookup error: xhc-hb04: undefined symbol: _ZN7IniFile4FindEPKcS1_i
xhc-hb04-world-mode.hal:2: xhc-hb04 exited without becoming ready
Shutting down and cleaning up LinuxCNC...

Any solution for that?

garymcrobertpdx wrote:

But I gather this time I will have to get the 2.6 source code and replace the IO routine with the
IoControl.cc you provided, compile, then proceed as before. Is this correct? If so where
do I find the source and I assume there is a make file to direct the compile process.

Not at all, as I said

I produced 2 binaries, plus the source

Use the binary in the later zip and replace /usr/bin/io with that (save the old file as io.old in case you need to revert)

If you needed to make other changes you would need to get the source, ( or you might just be able to compile it with comp, haven't tried) but this should work as is for master on 10.04

regards

OK but I am not clear as to the procedure. I think last time I used the build bot to create 2.6 rt
then Pncconf to create my_LinuxCNC_machine. After that all I had to do is add a few lines to
the LinuxCNC_machine.inii file to provide some path info and enable remapping of M6.

But I gather this time I will have to get the 2.6 source code and replace the IO routine with the
IoControl.cc you provided, compile, then proceed as before. Is this correct? If so where
do I find the source and I assume there is a make file to direct the compile process.

Sorry about all the newbie questions I do this sort of thing so infrequently that I fail to remember
all the details as to how to go about it. I will take notes for the next time I have to do this.

The bit file is indeed for a 200K 7I43. I can make a 400K one tommorow

I originally copied the files as root (have to be to copy) and I copied them to the 7i43-4 directory. Today I looked again and found a 7i43 directory in addition to the 7i43-4 as I did before. So I copied the files to the 7i43 directory and it now finds the file, must be looking in the generic 7i43 directory. It now says that the firmware is for a 200k fpga. I do indeed have a 400k fpga, there was some confusion in the beginning of my post as to which FPGA I had. So is the new file you made for a 200k or 400k fpga? It appears to be for a 200k.

-Scott

[ 1221.548165] hm2_7i43: loading HostMot2 Mesa 7i43 driver version 0.3
[ 1221.550359] hm2_7i43.0: firmware: requesting hm2/7i43-4/SVRM6S.BIT
[ 1221.552634] hm2/hm2_7i43.0: board has FPGA '3s400tq144', but the firmware in hm2/7i43-4/SVRM6S.BIT is for FPGA '3s200tq144'
[ 1221.552646] hm2_7i43.0: board at (ioaddr=0x0378, ioaddr_hi=0x0778, epp_wide ON) not found!

HostMot2 firmware goes in /lib/firmware/hm2/cardname/
You need to copy as root:
sudo cp some_old_bitfile.bit /lib/firmware/hm2/7i43-4/

It looks like you may have the file name in the wrong case
(the bitfiles I made had all lower case names)

PCW, thanks for compiling and testing the new file. Do I have to do anything after copying the files to the firmware/hm2/7i43-4/ directory? Linuxcnc says that it can't find the BIT file. I didn't do anything special last time I tried the original file you sent.

-Scott


Print file information:
RUN_IN_PLACE=no
LINUXCNC_DIR=
LINUXCNC_BIN_DIR=/usr/bin
LINUXCNC_TCL_DIR=/usr/lib/tcltk/linuxcnc
LINUXCNC_SCRIPT_DIR=
LINUXCNC_RTLIB_DIR=/usr/realtime-2.6.32-122-rtai/modules/linuxcnc
LINUXCNC_CONFIG_DIR=
LINUXCNC_LANG_DIR=/usr/share/linuxcnc/tcl/msgs
INIVAR=inivar
HALCMD=halcmd
LINUXCNC_EMCSH=/usr/bin/wish8.5
LINUXCNC - 2.5.2
Machine configuration directory is '/home/lathe/linuxcnc/configs/lathe2'
Machine configuration file is 'lathe2.ini'
INIFILE=/home/lathe/linuxcnc/configs/lathe2/lathe2.ini
PARAMETER_FILE=linuxcnc.var
TASK=milltask
HALUI=halui
DISPLAY=axis
Starting LinuxCNC...
Starting LinuxCNC server program: linuxcncsvr
Loading Real Time OS, RTAPI, and HAL_LIB modules
Starting LinuxCNC IO program: io
Starting HAL User Interface program: halui
Shutting down and cleaning up LinuxCNC...
Running HAL shutdown script
Killing task linuxcncsvr, PID=2317
Removing HAL_LIB, RTAPI, and Real Time OS modules
Removing NML shared memory segments
Cleanup done

Debug file information:
Can not find -sec MOT -var MOT -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
insmod: error inserting '/usr/realtime-2.6.32-122-rtai/modules/linuxcnc/hm2_7i43.ko': -1 Unknown symbol in module
lathe2.hal:9: exit value: 1
lathe2.hal:9: insmod failed, returned -1
See the output of 'dmesg' for more information.
2317
PID TTY STAT TIME COMMAND
Stopping realtime threads
Unloading hal components

Kernel message information:
[ 1020.019317] I-pipe: Domain RTAI registered.
[ 1020.019325] RTAI[hal]: <3.8.1> mounted over IPIPE-NOTHREADS 2.6-03.
[ 1020.019328] RTAI[hal]: compiled with gcc version 4.4.3 (Ubuntu 4.4.3-4ubuntu5) .
[ 1020.019335] RTAI[hal]: mounted (IPIPE-NOTHREADS, IMMEDIATE (INTERNAL IRQs DISPATCHED), ISOL_CPUS_MASK: 0).
[ 1020.019338] PIPELINE layers:
[ 1020.019341] fca5ce20 9ac15d93 RTAI 200
[ 1020.019344] c085cb20 0 Linux 100
[ 1020.045867] RTAI[malloc]: global heap size = 2097152 bytes, <BSD>.
[ 1020.045981] RTAI[sched]: IMMEDIATE, MP, USER/KERNEL SPACE: <with RTAI OWN KTASKs>, kstacks pool size = 524288 bytes.
[ 1020.045987] RTAI[sched]: hard timer type/freq = APIC/12500308(Hz); default timing: periodic; linear timed lists.
[ 1020.045991] RTAI[sched]: Linux timer freq = 250 (Hz), TimeBase freq = 3200599000 hz.
[ 1020.045994] RTAI[sched]: timer setup = 999 ns, resched latency = 2944 ns.
[ 1020.046069] RTAI[usi]: enabled.
[ 1020.120059] RTAI[math]: loaded.
[ 1020.206542] hm2: loading Mesa HostMot2 driver version 0.15
[ 1020.211398] hm2_7i43: loading HostMot2 Mesa 7i43 driver version 0.3
[ 1020.213584] hm2_7i43.0: firmware: requesting hm2/7i43-4/SVRM6S.BIT
[ 1020.215304] hm2/hm2_7i43.0: firmware hm2/7i43-4/SVRM6S.BIT not found
[ 1020.215309] hm2/hm2_7i43.0: install the package containing the firmware.
[ 1020.215317] hm2_7i43.0: board at (ioaddr=0x0378, ioaddr_hi=0x0778, epp_wide ON) not found!
[ 1020.861051] hm2: unloading
[ 1020.984066] RTAI[math]: unloaded.
[ 1021.044066] SCHED releases registered named ALIEN RTGLBH
[ 1021.060835] RTAI[malloc]: unloaded.
[ 1021.160024] RTAI[sched]: unloaded (forced hard/soft/hard transitions: traps 0, syscalls 0).
[ 1021.168086] I-pipe: Domain RTAI unregistered.
[ 1021.168097] RTAI[hal]: unmounted.

HI

There are 2 zips on that page of the thread.

Because some hard coding of paths goes on, depending upon whether the build is a RIP or /usr install, I produced 2 binaries, plus the source.

If building from source, delete /linuxcnc-dev/bin/io and substitute the source for /linuxcnc-dev/src/emc/iotask/ioControl.cc

This version was to suit tld70's toolchanger, but sounds like it should work for you

There are 2 new pins, iocontrol.0.current_tool and iocontrol.0.tool_altered
current_tool receives the number of the tool which has been manually loaded or otherwise altered
tool_altered is the trigger to prompt a tool change, which if the toolchanger holds the same tool number as current_tool, (it should do ,it sets it!), the tool number is updated without a change

The other wrinkle with this version is that it automatically updates to current_tool when first loaded.
This sets the tool number in Axis at start up.
If your toolchanger is not going to be setting current_tool straight away and you are manually doing it through a panel or something, that will need changing.

Normally both these pins would be connected to the tool changer component, albeit they could connect to pyVCP widgets or anything else that can provide the right tool number and trigger an update.

See how you get on, changing the startup behaviour is trivial if that needs doing

regards

Thanks for the fix to the software!

I have looked at the discussion in the thread I see the ioControl-2-6-mk2.zip file.

Is there a doc to guide me? I think I have done this process before but enough
time has passed that I do not remember how to go about it.

Chris,

With the onscreen keyboard can it come in below the MDI command box? With it being above my hand covers the line i am typing.

I also dont get the keyboard on the tool edit or offset pages.

Could you also make it so the keyboard button is selectable. It doesnt change i cant tell when the keyboard is shown or not shown. The button doesnt change it only works as a highlight.

The spare button on the side panel can you make that a calulator?

When running a program it would be nice if program edit was avaible when feed hold is selected.

When feed hold is selected cycle start should also be an option to start the cycle back. It was not clear how to start a program feed hold was selected.

Can you add a seperator bar or something on the tool edit page between the diameter and comments section?

When pressing enter on the run at line dialog I think the dialog box should go away.

tjamscad wrote:

cmorley wrote:

I pushed the increments fix, changed to the new layout of offsetpage widget and move the onboard keyboard to the center tab.

Chris M

What is changing about the offset page? I was also asked what the 5x on the offset page is for?

Mostly the layout. Axes are across the top. systems are the rows. (opposite of before but consistent with other tables).
You can now edit the systems directly rather then just G5x (the current system). The current system is in red.
Also hiding rows or columns actually works properly.
And finally there is a column editable for description. Eg if you of the systems is for the second vice you can name it 'Vice 2' as a reminder.

Chris M

tjamscad wrote:

cmorley wrote:

Can you post the INI file.
I bet you didn't add the jog increment entry to the INI (TOUCY doesn't use it)

As for jogging - did you remember to click the on screen button 'jog' to enable jogging?

Chris M

You are correct once again. No "INCREMENTS" in the ini file.

This should be why on screen jogging didn't work too.

To get the tool number updated at startup, I have changed iocontrol, which is what tld70 and cncbasher are using.

If you want to use the modified iocontrol to do it, a rip build version and install build version for 2.6 are on this thread
www.linuxcnc.org/index.php/english/forum...-tool?start=30#28977

regards