Advanced Search

Thank you for all the help so far. I've been running into an issue and feared it may have been caused by my fumbling earlier on and some weird state I may have put the installation into. To resolve this, I started a fresh install from current ISO and so far it has been much smoother given that I have a little more understanding. However, I have still ended up getting stuck at the same place, lol. I have included a screenshot and some config files to assist troubleshooting. 

Thanks, 

Steve Cap 

joint.N.amp-fault-in IN BITShould be driven TRUE if an external fault is detected with the amplifier for this joint.

Hmmm... Can't figure out how to edit a post. Here's an image of the controller

Hi all. I am trying to figure out the last step of configuring my machine. For background, I am using a Raspberry Pi 4b with the current LinuxCNC image from the downloads section. The RPi is connected to a Byte2Bot parallel hat which is in turn connected to a parallel port breakout board. I have managed to figure out all of the step, direction, and limit switch stuff and now have a machine that moves, homes, etc.

The final thing I want to do is define an input to LinuxCNC to stop motion if one of my closed loop stepper motors throws a fault. Essentially like hitting the E-stop. The stepper drivers each have a fault output pin and all of these fault outputs from the drives are wired to parallel port pin 10 of the BOB. This should be GPIO 25 (GPIO connector pin 22). For everything else I have done I have used templates and just been able to change pin numbers. I believe the hal entries would look something like this:

net driver-fault <= hal_gpio.GPIO25-in
net driver-fault => iocontrol.0.emc-enable-in

I believe this would result in a stepper driver fault essentially tripping the emergency stop.  I don't have a way to test this right at the moment and was wondering if someone could take a look and see if my thinking is correct?

Thanks,

John B 

Greetings.

I'm new to linuxCNC (and CNC in general). I've been "gifted" a Stinger CNC after the owner upgraded to a newer machine because he was having problems with the software on this one. It came with an OLD tower PC running WinXP and WinCNC. This was connected via parallel port to another box that housed the controller and other electronics like the emergency shutoff switch, etc.  The controller it uses is a Testra 4x483Q.  I've purchased a mini-pc with a parallel port and loaded linuxCNC on it. I've begun trying to get it configured to talk to the controller but it's not natively compatible and it's been giving me some issues. Frankly, I'd just as soon put new controllers into the box that are natively supported by LinuxCNC but before I go spending hundreds of dollars on what would be a stab in the dark for me, I thought I'd ask for some advice.  Oh, I've also purchased a 3-phase, 220V (chinesium) spindle with an H100-series "High Performance Vector Control Inverter" to replace the router it came with so I'd like/need to be able to control that as well.  Any help or advice you can offer will be appreciated. Do I keep fighting with getting the old controller to talk to LinuxCNC or do I get a new set of controllers, power supplies, etc that are known to work with LinuxCNC and build it back up myself (and hopefully learn something along the way?). Recommendations for controllers and power supplies would be great.

Thanks!

Thanks PCW.

At the moment all that’s in the kernel command line is quiet
I had previously tried to add an isolcpus=2, 3 using grub-Customizer but it messed up somehow and was forcing it to boot into tty which took a little time to work out and resolve.
I will add the power management in the file manually tomorrow and see if that resolves it. The pings are all low but as you say something must happen that causes some latency every so often

TangentAudio wrote:

I'm not surprised I misinterpreted the intent of that pin.  The current state of in-drive homing code and documentation makes it a bit of an "adventure!"

Yes, it is confusing. I tried to use joint.N.is_custom-homing to trigger the cia402 homing input but hal complained about the pin type so had to add a pin of type IO (joint.N.start-homing). I think I could sense homing is done when it is set to zero. But I am getting ahead of myself

Good evening Mr. PCW, everything is fine, I managed to increase the speed, thank you very much. The results of the electronic boards from the manufacturer Mesa were surprising, being able to increase the speed beyond what the machine can support.
Please, can you tell me which HAL configuration component I can use to connect MAX_VELOCITY and MAX_ACCELERATION with a variable in the classicladder or using a UI grid graph in the GMOCCAPY graphical user interface tabs? The machine operator requested to be able to increase and decrease the machine's operating speed without having to change the INI file.

dkms builds the drivers from source so you need linux-headers installed for the kernel version you are using. but first you need to
sudo apt purge r8168-dkms

then reboot, then
sudo apt install linux-headers-$(uname -r)
sudo apt install r8168-dkms

currently the "+" key in combination with a keyboard jog key will run at max speed, the "-" key in combination with a keyboard jog key will run at 10% of the jog speed slider adjusted speed. so you do have access currently to speed overrides with the keyboard in probe basic.

I ran a linuxcnc 2.9.2 instance with axis in a virtualbox. I look at linuxcnc status.

One call to G10 L1 P1 X10 and G43 works fine. g5x_offset don't move, tool_offset is changing to X10.
The same test with qtaxis needs two call to G10 and G43 in order it works.

I managed to do my offset change on the fly with a numbered variable and two calls to G10, so it's not a problem anymore for me, but I think there's a bug to look at here.