Advanced Search

Not sure how far behind the bookworm package is but '/sim/axis/remap/manual-toolchange-with-tool-length-switch' runs for me on a 2.9 build from February:

 

Maybe updating from the linuxcnc.org sources would help:
linuxcnc.org/docs/2.9/html/getting-start...dating-linuxcnc.html

One possibility that comes to mind is to
1. smooth the tool position trajectory (ie XYZ) as Grotius has been doing
2. create a second 3d trajectory that is offset from the tool position trajectory by a fixed distance along the tool orientation vector
3. smooth this second trajectory using the same method as in step 1
4. calculate the smoothed tool orientation vectors
5. calculate the smoothed orientation angles (ABC) from the smoothed tool orientation vectors.


I found a paper that may describe the method above but I do not have access to the full text:
www.researchgate.net/figure/3D-clothoid-...ector_fig3_361356637

The pins starting with 'FP4-gearbox.' would likely be coming from a custom hal component. I cannot say for certain as I don't know which hal file you got this from.

I'm a bit (or more than a bit) flumoxed at this point. I'm running a Dell Inspiron 3668 with a StarTech PCIe Parallel port card (AX99100x chip). I seem to be getting decent latency, but it doesn't appear to be working within the software. If I don't load the driver it complains that there doesn't seem to be a parallel port there, if I do load the driver:

sudo insmod parport_pc.ko io=0xd010 io_hi=0xd000 irq=14

then the error that no port there goes away (sort of). I have to sudo rmmod parport_pc.ko and then re-insmod it after each reboot. However, it still doesn't actually appear to be twiddling the bits on the parallel port, more disturbingly dmesg comes up with the following:

"parport_pc: loading out-of-tree module taints kernel."
"parport_pc: module verification failed: signature and/or required key missing - tainting kernel"
"parport 0xd010 (WARNING): CTR wrote 0x0c, read 0xff"
"parport 0xd010 (WARNING): DATA: wrote 0xaa, read 0xff"
"parport 0xd010: You gave this address, but there is probably no parallel port here!

So it really looks like I'm giving it the wrong address when loading the module, but they tell you to do an lspci -vv and it reports IRQ14 and gives two addresses 0xd000 and 0xd010. I've tried both combinations to no avail

Sorry didn't notice

 
 

The lathe screen is only available in the development version (2.10 master) of linuxcnc.

BIte the bullet and upgrade to 2.9.2
2.7 is just toooo old.. SO many new features

i am using this device as you recommended in this post
www.xt-xinte.com/DIEWU-EtherCAT-Slave-IO...Encoder-p748959.html

This may explain things. 
control.com/technical-articles/a-look-in...-npn-and-pnp-sensor/
I'm not sure if your device can use negative and positive as a common.

Also have a look at some other ethercat documentation for wiring diagrams

Are you 100% sure this board is suitable for Linuxcnc.
If it's for an embedded application, I not sure if RPi s are certified for Industrial use, would be a better option, I'm sure it's suitable for ethercat.

Maybe 3rd time lucky replying.

1 Development tools & libraries to not appear to be install.
2 There seems to be a lot of x86 Linuxcnc binaries in the image.
3 Once Linuxcnc has been built, you will need the runtime libraries to suit, if these aren't available it will be a hard task (as these may require building from source)
There is a utility to convert deb packages to OPKG formats, so this maybe an option.

Unless there is a way to get a version of Debian arm installed it may be a very very hard battle. But Debian does support ARM64.

Here's some info:
forum.linuxcnc.org/9-installing-linuxcnc...parallel-port#303645

There is always the chance that lspci will show the card ports as "disabled", but it's usually the case that drivers aren't required and are aimed at the serial part of the chip, these chips have the ability to work as a serial port, parallel or SPI port (this I have seen as a reference design and not in the wild).

Linuxcnc doesn't use the kernel driver to talk to the parallel port card. It has it's own driver that talks directly to parallel. The only real call the the kernel driver is to ask for "exclusive" use of that port. Even if that call fails and the port address you give Linuxcnc does exist it will merrily go on its way.

As mention in my linked post there are a couple of methods to work out the port address.

I knew this topic would pop again in a week or 2 after starting that thread.

how?? i want to connect two pins from VFD ( spindle forward and reverse) and do you mean the common is the power input of the group ?

The outputs are probably NPN which means your load needs to go between  the output pin and ground and common will need to be +24V

Hi Grotius,
I have been trying to find out the physical meaning of curvature and how to imagin G2 continuity in higher dimensions. I have not been having much success in that endavour.
This recently published paper I have found claims to use splines for filleting in 5 axis: A Novel Local Smoothing Method for Five-Axis
Machining With Time-Synchronization
Feedrate Scheduling ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9085380

I was hoping for some help to find some references on how to do filleting in higher dimensions to reduce the tangential jerk.

-automata