Advanced Search

Aciera wrote:

Of course we are interested! Please share.

x 2 !

I was going to suggest a pull request to github.com/linuxcnc-ethercat
That way it will find its way into the Debs for all to use....

Of course we are interested! Please share.

 

jogging operation is independent of the type of motors used. Have you actually confirmed in halshow that these pins are set correctly?

axis.L.jog-scale
axis.L.jog-enable
axis.L.jog-counts

Hallo, Mesa bietet das 7i76eu als Ersatz für das normale an.
store.mesanet.com/index.php?route=produc...83_85&product_id=399

Kann ich es genauso verwenden wie das normale 7i76?

vielen Dank im Voraus

I suggest you contact Treotham. They will be in VIC. Call them
treotham.com.au/
For stepper wiring on a couple of machines I have used ÖLFLEX CLASSIC 400 CP 4X0.75 from them
Then had to get it in from overseas but it didn't take long.They will have other options.
This was rated for drag chain use and actually has two layers of plastic sheath as well as the shielding.
To get wires up on the gantry, I use DB9 breakout boards and you can get premade cables from www.lindy.com.au/. These are pretty light weight but sure beats soldering up your own.
The rest of the (unshielded) cabling I mostly use Jaycar hookup wire with bootlace ferrules. You can  get a kit with the crimper and ferrules from eBay. The Jaycar wire is a bit light but it works...
 

Just for the record the machine is homed, but actively changing axis.x.jog-counts moves nothing.
Are the published the mpg hals I have adopted for stepper motor systems, and not appropriate for my DC servo system?

Note that ' s-output => spindle.0.on' does not make much sense in the context of the hal sections posted above as 's-output' is a float signal and 'spindle.0.on' is a digital out pin.

Looks like you are actually missing the input to enable the stepgen.

So in the spindle section of your hal file change this line:
to this

Tried to follow how this works in the onCircular function.
Three lines down there is a test with 0.2 mm line 452. That fails for this circle.
Then ends up doing a new circle line 476, which is the one that isn't really right.
I don't understand this with merging circles but I guess it has something to do with that.
Edit: In the fusion simulation of the tool path it is clear that a circle is made up of several circle segments, which is then supposed to be assembled to a full circle in the post-processor. Guess something went wrong with the assembly.

Hi, there,  I am trying to follow step in this thread to set up stepper as a spindle. But the spindle stepper is not turning. I have tried to add this line as you mentioned. The LinuxCNC has a  problem with s-output  => spindle.0.on. Do you have any idea what I should have fix or add to hal in order to work properly. Thank you very much in advance.

My Gentoo installer+image has been published. I tuned it as much as possible, I don't think Clear Linux will be much faster than what I have, if not slower.

github.com/NTULINUX/gentoo_backup/releases/tag/v0.3-alpha

openlunchbox.com/mw19/index.php/HOWTO:_Gentoo_for_LinuxCNC

If you're using the installer on Bookworm, start here:

openlunchbox.com/mw19/index.php/HOWTO:_G...tching_the_installer

To tell the truth for connection of step & direction from my 7i76 to the drivers I didn't bother with shielded cables.
When I get around to the limit switches I may use shielded cables, I don't know.
But what I was very particular about was the path of the power cables.
AC power was physically isolated as the installation allowed, at no instance would I allow AC power to cross any signal wire.
At the same time DC power was physically isolated from the AC power and any signalling.
The basic plan was AC/DC run on one side of the cabinet and signals on the other side.
What made my install easier in the above is I used an 18 unit rack cabinet, used upright rails in the middle of the cabinet, thereby giving me twice the area to install gubbins, AC power was assigned to one side of the partition, think relays auxiliary power and such, and the PC Mesa cards and such on the other side all mounted to metal plates.
Another thing I employed, maybe over the top, is using a seperate power supply for powering the relays and contactors that regulate the AC power supplies used elsewhere. All power supplies are controlled via a No Volt Return setup, as I'm lazy I made some PCBs to make the wiring easier, so if I do lose power, thanks to Labor's green madness, the power doesn't come back on and suprise me.
Meanwell & Wiedermuller PSU are used throughout, the servo drive that Powers the AC spindle motor has been installed with the required AC filters and line reactor.
So far I've no issues with interference either from the mill and mill electronics nor with any appliance in the house.
Ok back to your original question, I would have a look on AliExpress for cabling, they seem to have a large range. Some of our local stuff, Jaycar Altronics et all are kind of expensive.
Whatever your do use bootlace ferrules, good quality crimp connectors, do not solder the crimp connectors. A good quality crimp cold welds the wire to the connector, same as wire wrapping. The solder has a tendency to whick up the wire and create a "solid" point where with the right amount of movement it will fracture. Forget what any "YouTube" guru says about soldering they are wrong. Buy good quality crimps and crimp tools, there's a company in Engadine NSW Ive bought all my stuff from and never had an issue.
Ok I come back with some photos and such. Bear in mind this is for a mill install, a plasma setup will have more requirements, Tommy or Rod hopefully will chime in .

I've probably forgotten way too many things as well.

Hi guys.
For my work we are controlling robots with the use of ek1100 and two el6751.
I made a driver together with etherlab where you can use linux native can0 and can1 to control can devices.
I control multiple servo motors.
Are there any developers who want to know the design and how I did it. Also there is a bug in the EL6751 eeprom, took me a long time to find it out. With my fix and driver native linux can communication with el6751 is easy.

Regards Grayson

 

Can all this operation of unrolling to be transparent to gcode?
For example to have a gcode of cutting
diagonally a sheet of 10m.
plasma if y axis is 2.5m must start and stop
4 times.
Can this handled automatically without specific gcode?

The image has finally been posted:

github.com/NTULINUX/gentoo_backup/releases/tag/v0.3-alpha

Do not use the tarball directly as it meant to be used with the installer.

For some reason, the axis interface doesn't work due to this bug:

github.com/LinuxCNC/linuxcnc/issues/3205

I previously had qtdragon working, however, Gentoo is moving away from Qt5 so all GUIs that depend on Qt5 will not work. I highly recommend using gmoccapy unless you want to help me figure out why axis doesn't work.

I'll be working on an RTAI image soon, but for now, only PREEMPT_RT is supported.

Instructions:

openlunchbox.com/mw19/index.php/HOWTO:_Gentoo_for_LinuxCNC

Please note, I switched the checksum from SHA-1 to BLAKE3/b3sum, so you must have b3sum installed if you are using the installer on Debian. I don't think SystemRescue comes with it either but it is much faster and more secure to use BLAKE3 over SHA-1.

FlexGUI also works, so if you use it, you're in good hands. Turboss and Lcvette (I think it's Lcvette) is working on porting qtpyvcp to pyside6 (Qt6) so qtpyvcp should be working in hopefully the near future.

If you're using Debian, you can start the guide from here:

openlunchbox.com/mw19/index.php/HOWTO:_G...tching_the_installer