Advanced Search

Part of the hole processing smarts is in the qtpyvcp code base.  In the programe used by monokrom to parse out the incoming gcode file and adjust it if necessary.  I still want to come back to this and refine/improve it.  Just on a long list of other things in life to see to.

tommylight wrote:

This time you can blame it on me as if i recall correctly, Parallels is the software running on Apple hardware and makes it possible to run other operating systems in it, in short running virtual machines.
If that is correct, then all my answers are on point, it should run any ISO without making images on other hardware, that is the whole point of it.
Then again, it is Apple, so maybe only runs ARM stuff?
 

Both links I included explain in explicit detail why I need someone with an Intel Mac and Parallels to create a VM on that hardware for me load into my Apple Silicon Mac. It was, and still is, all that I need to try out x86 emulation in Parallels 20.2.

Changing the Z-axis accel only slightly affected the rapid distance, if at all.

The earlier tests were done with MAX_VEL = 4, and MAX_ACCEL = 175.

Dropping down to MAX_ACCEL = 2 resulted in approximately the same pre-synch rapid distance at all speeds.

Much less violent rapids, of course.

Update here:
forum.linuxcnc.org/38-general-linuxcnc-q...blem?start=10#319865

So my lathe's behavior is closer to the OP's when running a program vs. MDI.

But as I mentioned in the other thread, I don't think this behavior is correct even with the reduced 'dive' towards the synch point.

Thanks for the troubleshooting suggestion.

Yes, same behavior, but much less Z-distance during the initial 'dive' than when using MDI.

Approximate position the Z-axis rapids to before synchronization while running a program:
G33.1 Z-0.5 K0.05 $0
(Starting at Z0.500")

• 60rpm: 0.440"
  • rapid dist: 0.060"
• 100rpm: 0.390"
  • rapid dist: 0.110"
• 200rpm: 0.370"
  • rapid dist: 0.130"
• 400rpm: 0.300"
  • rapid dist: 0.200"
• 750rpm: 0.140"
  • rapid dist: 0.360"
• 1000rpm: -0.125"
  • rapid dist: 0.625"

So perhaps 1/4 the distance vs MDI.

Still don't think this behavior is right.

Have you tried menu-config to configure the wireless network ?
will show whats happening with the wireless network during connection, then you can use google to search for any issues you see, eg associating, disconnection and the like. As this will be a general Linux issue.

gene_weber wrote:

Rather than starting a new thread about jitter mitigation, I'm just going to add it here.

First, I'm keeping in mind that Peter Wallace said " The jitter is basically unimportant with a 7C81 unless its so bad you
get real time errors at the normal servo thread rate."

I'm also treating the Raspberry Pi + 7C81 as a dedicated controller, not a PC. By that I mean I'm not going to be doing anything on the Pi except LinuxCNC. I may turn on WIFI to update packages, or use email to transfer a file. But I plan to even turn off WIFI before I launch LinuxCNC.


I ran several tests to see which combination of settings offered the most optimized jitter. I ran two Glxgears for all tests as I though that was reasonable, and ran the tests for long periods of time unless they looked bad early on.

I thought that turning off all unused services would decrease Jitter. So I turned off 9 of the 23 services, such as CUPS, Bluetooth, etc. To my surprise this did not decrease jitter.

Screen blank increased jitter a bit, so I turned that off. It doesn't make much sense anyway on a CNC controller.


These are the settings I added to the line already in /boot/firmware/cmdline.txt:

skew_tick=1 kthread_cpus=0-2 irqaffinity=0-2 rcu_nocb_poll rcu_nocbs=3 nohz=on nohz_full=3 isolcpus=3

I tried a few others that some people recommended on various blogs. Nothing else I tried had any noticeable positive impact on jitter, and some noticeably slowed down the responsiveness of the Pi.


Turning off timer_migration did reduce jitter. What I read said that you could make this setting permanent by adding it to the boot command line, or setting it in either of two different system files. Oddly none of that worked. So I setup a cron job to disable it at boot, and that works.
[code]crontab -e
Add the following line:
[code]@reboot sudo sysctl kernel.timer_migration=0


The other thing that helped reduce jitter was changing the scheduling policy to real-time and increasing the priority. I decided to create a shell script to launch latency-histogram, and eventually linuxcnc, which changes these things on the process.

[code]#!/bin/bash
#
# Run with the SCHED_RR real-time policy at the highest priority level (99)
# Uses chrt to switch the process policy and priority immediately after launch.
#
# Gene Weber - January 2025
#
# "The difference between SCHED_FIFO and SCHED_RR is that among tasks with the same priority,
# SCHED_RR performs a round-robin with a certain timeslice; SCHED_FIFO, instead, needs the task
# to explicitly yield the processor." -Claudio

priority=99

# Truncate $1 (the command) to 15 characters starting at 0.
command=${1:0:15}

# Execute the command line provided to rrt as a background task.
`$@` &

# Get the process ID of the command.
PID=`pgrep $command`

# Change the real-time policy to SCHED_FIFO with a priority of $priority.
sudo chrt -r -p $priority $PID


I named it rrt (run real-time). So to launch latency-histogram it's simply:
[code]rrt latency-histogram --nobase --sbins 1000

All spawned sub-processes inherit the policy and priority. So if linuxcnc is launched with rrt:
[code]rrt linuxcnc
all of the processes that are part of LinuxCNC have the policy of SCHED_RR and priority of 99. I used Round Robin scheduling to make sure no LinuxCNC process "starves" another.

I don't know if this will work in the end, but I'm going to give it a try. Nothing ventured, nothing gained.

Implementing all of these, here is the jitter histogram. Good enough.
[attachment=67004]final-16Jan2025-3117.png[/attachment]
 

[/code][/code][/code][/code][/code]

 

I think it's just easier to isolate 2 cores, it really doesn't affect performance. And reading the docs, mainly pertaining to amd64, the general rule is to isolate cores on the same physical die, for reason beyond my pay grade.

One thing that might be worth of note, as development moves along, is to check the Makelfile in the top of the directory as minor revisions can change for the the rpi-linux kernel.

Or you can run

Analog voltage from inductive sensor? Is that the position sensor for the ram/cylinder?
Maybe using the values from that sensor to home or to check and retract the ram at LinuxCNC start?
Also use the value from that sensor to stop the machine from moving anything else if not in upper position. I would strictly tie this to motion feed hold if this is punch press so the part can not move until the ram is up. This i think is a must on such machines. Also to make sure the material has stopped before ramming down...

Getting a Hunk #1-9 succeeded, and a few lines later (right after trace.c) a HUNK #1 failed during the patch. 

install and kernel activation does seem to complete if it’s ignored. 

Is the cylinder controlled by -10 volts advances and +10 volts retracted position 0 volts gradually descends
Note: I am using translate google sorry for any inconvenience

Thanks for the feedback, good suggestion, but since I'm very late on this project, physically installing a sensor will take more time and I don't know if the client will accept it on the next visit. Your idea helped me to try a different way, maybe never tried before in LinuxCNC, which would be to take the analog voltage reading from the inductive sensor that is present and already part of the original machine and create a logical ladder to compare the voltage from zero to 10V where 8.6 volts would be the virtual limit switch. With your experience, do you think it is possible to use an analog reading and determine certain thresholds to turn on and off a bit memory in the ladder and use it as a limit switch?

Found my issue. The internal encoder parameter is not getting set properly on the 7041's In my XML config file I have the encoder modparm set to false but when i open a terminal after axis has started up and check the parameter using "ethercat -p upload --type uint8 0x8012 0x08" it shows it as zero instead of 1. I suspect something in the driver code is not jiving with the revision of the modules i have. When I set the parameter manually after axis has opened up. The simulated encoder feedback is updated on the pins as it should be and all is well.

Is the cylinder controlled by +-10V or just 0-10V?

Is it setup as a normal axis?
Adding a switch and using it for homing would work in that case.
Adding the switch would be advisable in any case and set it in hal so it does not allow anything to move if the cylinder is not in the upper position.