Advanced Search

I don't understand the problem here, really.

But with 1600 pulses per rev and a 3:1 belt ratio, the scale is just 1600/3 = 533.3333333333333333333333

If you see a creep, it could be the missing threes.

But, LinuxCNC expects pulser per _degree_ for a rotary axis, so you might actually want 1.481481481481481

Actually, hal_gpio might "just work" in this case.

Try this command and see if it reports the pins:If you get "command not known" then
If the pins are reported, then:
linuxcnc.org/docs/stable/html/drivers/hal_gpio.html

Note that unless the index pulse is from the spindle shaft, or the gear ratio is exactly 1:1, you can't peck-tap with the encoder on the motor shaft. This probably isn't a problem in most cases, but is worth being aware of.

There is a fixed but parametric G-code file as part of the config that does the feeding / multiple passes.

You should definitely look at that file to see what it does, and also what it assumes. The system assumes that the tip of the cutter is touched-off to the workpiece radius (ie Z = 0 is on the work axis)
Also note that the axis assignment is a little unusual, with hobbing radius as Z and work axis along X.

"shift" is a profile shift, simply a radial offset of the hob relative to the workpiece centreline.

Hi,
thanks for the help.
I've been messing around with the project but can't seem to get it to work.
I compiled ethercat_twincat and ran it but it's not reading any ports from my ethercat device. There are no input/outputs. Is there somthing else i need to configure?
My device is in PREOP mode and erroring because of:
"The current and the projected module list do not match. The data
exchange is still in progress." or
"The gateway is still in INIT status, the module list has not yet been
written to the gateway."

I should be actively writing something to the gateway to configure it i guess...

I would suggest that you connect iocontrol.0.tool-prepare to the turret-unclamp output pin, and then connect carousel.0.enable to the turret-unclamp-ready pin.

That way a tool change request will immediately unclamp the turret, and the carousel component will only start up once the turret is actually unclamped.

Probably, some experimentation may be needed, especially regarding how the re-clamping happens.

For the benefit of anyone else using the L510 (or the RS-branded RS510) VFD this component seems to work:
github.com/gitcoffey/l510_vfd
after making this change
github.com/gitcoffey/l510_vfd/pull/1
Install with

I assume that you are using the tool-number in HAL rather than pocket-number to bypass the tool/pocket swapping inherent to a random TC?

Is it possible that you could save a copy of the tool.tbl file before and after the problem and attach the two files here? That might help work out what is going on.

Check in hawshow if there is an enable pin for this window glade ...

If you have a different theme color, you can fix it by looking at the last comment on this post.

forum.linuxcnc.org/gmoccapy/45856-spindl...capy?start=40#313268

Hi, I'm new to LinuxCNC and I really need your help on this issue I run into.I created a custom button in Glade.
So I made a Top Level window named window1, put in a HAL table, hal_table1 and inside the HAL table a HAL button, hal_button1. I save the file as test.glade.
I'm using Glade 3.8.6 on LinuxCNC 2.8.4.
I added to the .ini file the following lines:

- In AXIS:GLADEVCP = -u hitcounter.py test.glade
POSTGUI_HALFILE = andrei.hal
In andrei.hal I have: setp gladevcp.hal_table1 TRUE

The button is show on the interface and it's active. This is working perfect.Now I want to implement the same in gmoccapy

-In GMOCCAPY:
EMBED_TAB_NAME = test
EMBED_TAB_COMMAND = gladevcp -c gladevcp -x {XID} test.glade
EMBED_TAB_LOCATION = box_custom_2
POSTGUI_HALFILE = andrei.hal
If I comment the setp gladevcp.hal_table1 TRUE in andrei.hal, the button is shown in the interface, but it is greyed out.
When I uncomment the line in andrei.hal the LinuxCNC is not starting and I have the error:andrei.hal:1: parameter or pin 'gladevcp.hal_table1' not found. Please see the linuxcnc.txt error report.

If I comment the line in andrei.hal and start the LinuxCNC with Gmoccapy interface, the button is greyed but I see the gladevcp.hal_table1 pin and I can send in halshow the setp gladevcp.hal_table1 TRUE command and it's working fine.
Also after loading the interface, I can see pins with halcmd show pin | grep hal_table1.

So as far as I can understand, the gladevcp.hal_table1 pin is not available at runtime in gmoccapy.
I tried a lot of settings with these commands and configurations and I don't know what I'm doing wrong, as aI follow the exact guidelines from the documentation.

I attach the .ini file. test.glade, andrei.hal files and the eror report linuxcnc.txt.

Presumably this is in C?

You may need ti #include A.h in B.c

Whole bunch of answers here and the appropriate place to discuss it.
forum.linuxcnc.org/9-installing-linuxcnc...official-images-only

You'll have to give a description of your setup, what image are you using, what is the latency & ping times.

Maybe this helps:
linuxcnc.org/docs/html/gcode/overview.ht...de:polar-coordinates

Hello, unfortunately I don't have a contribution to your CAM system, but I do have a question about the Raspberry Pi 5 and Linux CNC On the current PC I have problems with the latency, especially in the Ethernet connection to the 7i96s. Unfortunately, this always leads to unexpected stops of the machine. The last PC was no better and I don't want to buy a third one that will cause problems. My hope is that the Raspberry Pi 5 will work easily and without problems, even if it is a bit more expensive than a second-hand PC. Replacement would also be possible at any time. Now to the question What are your experiences with the pi 5 and which version are you using? Which SD card? Any special settings? Thank you very much Greetings Erwin 

Thanks for the idea. When I ran a few simple tests with no effect. I'm not sure how to test for emi. I expect if there was some my test would have had a reaction change but nothing changed. I moved the VFD as far away as I could and added a grounded makeshift shield.