Advanced Search

The BIOS give me the option to limit all PCI E to gen1 speed. There is no option/feature like XMP/EXPO option, but one to set the ram speed manually, so i set it to a lower one. 
I also tried to use different pci slot for the 6i25 (for the graphics card, there is only one slot available).
I also have an M2 SSD installed, that also uses the pci lanes if i get it right. Tried to use a HDD with new live image installed instead.

But no luck. 

rtai on the other hand runs absolutely stable, i already drove more than 1km with the test equipment 

 

I am still working on the Hardware. Had a hell of a time with the M23 connectors and the data sheets for old Harmonic Drives. Should be able to get it working electrically this week. Next step is commissioning in TC3 (which I have never done) to check the config for validity.

Assuming the power supply (X?) is controlled by LinuxCNC, shouldn't
this all be in the gcode with the feedrate just setting the total time? 

 

Go leave some likes for the man, he has done a very nice job of explaining in detail why it is so good and why it lost to Java.

Hi,
did you make any progress?
The EL8 are not at all the same as the A6, so there should be another solution...

Which usb devices did you try? Some links, please.
A RS485 to usb device is just a serial device to linux and there are plenty supported out of the box.
I would be very surprised f you got two that aren't supported by linux.
Plug in both of them of show here the result of the two commands

In BIOS set the PCI-E that the 6i25 is plugged in to GEN1, save and reboot.
Also, other hardware can cause such issues on PCI-E lanes as main ones (usually 20 lanes) are on the CPU, so setting the graphic port to GEN2 or GEN3 might also help, especially if you have a GEN4 board with GEN3 graphic.
And since most errors mention memory, try disabling XMP/EXPO.

Thanks for your fast reply. i read some linux forum posts to and try to play again with some cpu bios settings, but with no positive result. trying a newer preempt rt kernel seems to be some additional work.
So i think switching to rtai is the best choice at the moment. 

Thanks again for your suggestions.

possible solution

forum.linuxcnc.org/39-pncconf/56464-mesa...thcad-pncconf-screen    

thnx kworm

Thanks Hakan,

I followed that thread and tried to test hy_vfd with via halrun but still got the same results. I know the adapter is good as I was using it with Mach3 however, it required a driver and a plug-in to work. I've tried to find a linux driver for the adapter without success.

I decided to purchase a new adapter from Silicon Labs. I've downloaded the linux driver and have attempted to compile and install it. This has lead to much frustration as not all of the <include> header files are in the expected directories. To resolve this issue I renamed the original /usr/include/linux directory to linux_orig and created a new linux copying all of the contents of the linux_orig directory to the new directory then adding the missing header file to new directory in their expected locations. This resolved some issues but surface others.

To resolve the issues, I've added an additional -I path to the /usr/src directory where all of the linux realtime headers reside to the Makefile without success.

Not sure I can resolve this issue at this point.

Does anyone know of an adapter that works with hy_vfd natively?

after editing the PNCCONF according to kworm - update 
forum.linuxcnc.org/39-pncconf/56464-mesa...thcad-pncconf-screen

is possible to configure the step-dir spindle in 7i95T

Could you post the G-code that is giving the error? Not necessarily the the entire G-code program. Especially if the program is large. Just a snippet that includes at least a few lines before the error, and a few lines after, so we might be able to test and reproduce the error. It's a little hard to see and understand what is going on just from your image.

The error joint2 limit error is not caused by your hal or INI settings.
It results as it tells you, because the movement in line 11 will exceed the soft limits.

As you mentioned, that will happen more often with long tools, as your code will use a length compensation for tools.
If the preview window is updated, you will see that limit violation too.

Try the following:
- Change to the longest tool.
- set Z axis to cero on the top face of your workpiece.

Now it should work. 
 

I did a complete re-install and made sure I selected the right board starting config, and now everything works as normal. I must have accidentally started with a tangBOB starting config instead of the TangNano9K, or something along those lines, so as I was re-installing linux/checking hardware I brought the problem along with the original JSON file every time I tried fault finding.

If nothing else I can serve as a bad example !
Cheers.
Russell.