Advanced Search

What I would do is get a basic working configuration with pncconf.
This means setting the machine limits, scaling, directions, timings,
velocities, and accelerations. Once this is working I would then worry
about tuning for best performance and  special features like drive fault
operation.

That is, don't get the sandpaper out when you are still in the chainsaw stage.

 

The real time thing in the kernel seems a little confusing at the moment.

Unless a change is made to the /arch/Kconfig file you are unable to select a fully_preemptible kernel, which would appear the one required for Linuxcnc, the available info refers to this as RTOS. I'm not exactly clear about the amd64 kernel, as ATM I've only been interested in arm64 and specifically the bcm2712 kernel. I have tried the change the above mentioned file and can confirm the behavior.

Unfortunately I don't have a free SD card for testing and unable to get out to obtain one. So I can't comment on the kernel when booted, which is frustrating as I like to test things before commenting.

There is also the ability to select the type of real time kernel with via the kernel command line.

I'm trying to be a bit more "socially acceptable" for the coming new year, but yes that would be the way to go.

It would seem it didn't go over your head.

Yeah, now it dawns on me that it already has scaling built in, from -1 to 1 i think, so that is sorted.
Pretty sure it can make do without it, but it is a nice to have.
Again, fuzzy, but roughly in hal
net shorted-move-back motion.adaptive-feed hm2_7i96s.0.inm.00.input-10
yeaaaah that wont work...
setp ....
dropping the towel...
Aciera, HELP
Sorry, but i really need rest, it is 3AM here.

Some great info there.
Thanks Andy.

Through luck and a Google search, I have found: www.forum.linuxcnc.org/38-general-linuxc...learpath-sdsk#112579

It doesn't mention HLFB in the title at all, which is why it took me this long to find it, but it looks like there is a clue about how to start the process of integrating it. Unfortunately, it's an editing of HAL step, and I still don't know where to start in pncconf. How do I assign these pins in pncconf?

Just start a new thread that is quite easy.
Have you a link to the kits that "don't offer access to the SD card" ?

As far as I have been able to see there has been no major hardware updates to the RPi5. Or are you referring to the ability to boot from NVME ?

When claims like that are made it would be nice to provide a link, it's rather frustrating for someone to mention something like that with no references, as this would be a major shake up for the RPi platform.

Does this patch allow you to select fully preemptible kernel, when configuring the kernel ?

From my experience building with the mainline rt "patch" it was required to edit kconfig under arch. Otherwise there were only 3 options for RT kernel, which didn't include the fully preemptible option.

PCW wrote:

Its likely you have only 1 encoder enabled in your hal file.

To check the pinout (and number of encoders):

mesaflash --device 7i96s --addr 10.10.10.10 --readhmid
 

Thank you getting somewhere......

IO Connections for P1 -> 7I96_2
Pin#                  I/O   Pri. func    Sec. func        Chan     Sec. Pin func   Sec. Pin Dir
P1-01/DB25-01          34   IOPort       QCount           1        Quad-A          (In)
P1-02/DB25-14          35   IOPort       QCount           1        Quad-B          (In)
P1-03/DB25-02          36   IOPort       QCount           1        Quad-IDX        (In)
etc......

This is how I previously tested the basic DRO concept with A&B driving isolated inputs 0&1:
loadrt conv_s32_float count=1
addf conv-s32-float.0 servo-thread

loadrt scale count=2
addf scale.0 servo-thread
setp scale.0.offset 0.0
# 0.02mm = 0.0007874"
setp scale.0.gain 0.0007874

net ydro-in-i hm2_7i96s.0.inm.00.enc0-count => conv-s32-float.0.in
net ydro-in-f conv-s32-float.0.out => scale.0.in
net ydro <= scale.0.out

So now in my HAL file what should I replace hm2_7i96s.0.inm.00.enc0-count  with when using the P1-01 & P1-02 pins instead of the slow but isolated TB3:1&2 pins?


Much appreciated
 

Maybe get a newer Debain version?
2.28 is the last version in debain 10
Linuxcnc 2.9 comes in Debain 12

Hi Tommy,

Thank you for the response.  I am not sure what is obvious right now. The pin I am trying to use is motion.adaptive-feed which is what Aceira suggested.  From the motion page it seems that when this pin is set to 1 the feedrate is what was initially set.  Then this value can be scaled.  Its a multiple.  So to reverse feedrate at the same velocity in reverse this pin should be set to -1. 

This is how I understand it but could be wrong.  But I just don't know how to do this. 

carousel will just control the motor turret and rear the sensor for turret position. you have a lot of other things that need to happen to effect a tool change.

which make and model of tool changer ?

www.sauter-feinmechanik.com/en/scheibenrevolver.php

Shooting from the hip here, but, having that pin .... oh cr@p, i need sleep badly, having a hard time keeping things organized inside my noggin, so here goes an incoherent train of thoughts:
-probably would need scaling from the GUI to set the reverse speed, i do not think there is a need to back off at full feed rate
-should be active only when the input pin is active (mister obvious! )
-can not recall what the pin is for reverse run....
-
What did you try? Maybe it jogs a few gray cells...

rodw wrote:

But as PCW says, seeing 0 volt and 45 volts indicates it is working.
 

As i understood, he had 0V before firing the torch, and after the first firing it drops to -45, that absolutely points to THCAD working prior to firing and fails miserably after the firing as be it THACD output or 7i96S inputs are gone from the huge amount of EMI the actual torch firing causes and THCAD being enclosed inside a metal case where the source of all that EMI is.
-
The THCAD should be wired at the rectifier, not at the plasma output.
And yes, i have a THACD10 wired directly at the output lugs on an inverter plasma, but i would never advise that.
-
Just to be clear, yes you can mount the THCAD inside the plasma source, but absolutely should be avoided by inexperienced members of the human race.
The plasma source has a plethora of warnings plastered all over it, and the manuals spend several pages describing just how dangerous those things are.

Hi all,

I have given up on trying to have a PID control of the feedrate.  Instead I am using an arduino to set an input pin (hm2_7i96s.0.inm.00.input-10) to TRUE when the signal from the EDM power supply shows a short.  I am able to confirm the pin is going TRUE using hal meter.  That is the extent of my success. 

I am trying to use this pin to set the feedrate into reverse when a short is detected using motion.adaptive-feed. However, absolutely nothing I do will work.  I was trying to do this inside of the .hal file and that did not work. I also tried to do it in a python script and the GUI refused to load when this was attempted. 

I really have no idea how to program linuxCNC and I am growing a little desperate and could really use some help. Could someone please walk me through how to use the input pin hm2_7i96s.0.inm.00.input-10 to toggle feedrate from forward to reverse.  I feel like this should be straightforward but I am seriously struggling.

Thank you,
-Dylan