Advanced Search

Hi guys, haven't been around in a while. Hope all is good with everyone.4

I have a customer that has an old system I did about 7 years ago and the machine has been moved. Getting a weird issue. With every turn of the ballscrew it jumps. With the halscope I can see the f-error and then it corrects itself and it jumps abpout 1/16 of an inch.

I am having a hard time trying to figure out what is causing this.IT is only doing it on the X and Y axis. The Z is perfectly fine.
If I disconnect the velocity command from the mesa 7i77 card and put a battery on the wires going to the drive it works smooth. This I am baffled with.

I would like to upgrade Lcnc to latest version that can update if required. IS the version 2.9 iso version the best way to go? Also the configs I dont really want to start over so how would I get the old config to work on 2.9 version. I think the old opne is 2.6 or 2.7.

Thanks for any ideas,

Bob

 

Probably not very elegant, but what I have done is:

install Raspbian on a SD Card
use the imager to install LCNC image from the LCNC.org downloads page
leave the boot order to boot from the SDcard preferentially
remove the SDCard
system boots from NVME into LCNC

So far, this has enabled me to keep the Raspbian SDCard handy to reinsert and reinstall LCNC quickly as I experiment with different things and muck up the LCNC install...

This strategy inspired by my innate lazy nature will probably come back to bite me eventually...

Randale

Thanks for the answer but this didn't help me. 
First of all, the gcode which lightburn generates is using M67. I then read thatNoteM67 will not function unless the appropriate motion.analog-out-nn pins are connected in your HAL file to outputs.
I don't have that in my hal-file generated by stepconf. So i probably need to connect that to some pins or signals or whatever and put it into the custom-hal. I just have no clue how to do that. 
I tried adding this : 
net analogout motion.analog-out-00 => pwmgen.0.value
net laserpwm pwmgen.0.pwm => parport.0.pin-01-out
but then linuxcnc is complaining about :
./custom.hal:3: Pin 'pwmgen.0.value' was already linked to signal 'spindle-cmd-rpm'
I attach my latest hal and ini files.
 

The kerf does not change that much. Your typical kerf is around 0.2mm and even in lets say 20kW and 40mm thickness is might be 0.4mm.
Maybe you want to add the nozzle type as a different nozzle would require different parameters.
And you are correct about the focus and cut height, these are 2 separate things. One is the focal point of the laser ”dot” and the other is a physical height of the cutting head.

That was just to prove a point, keep chugging along, the experience you get from it is priceless.

When I google change boot order of RPi5, all links say use rpi-eeprom-config --edit and put the 6 at the RH end of the BOOT_ORDER line.
No other option is offered.
Why shouldn't it work, the update page I was pointed at says its for Pi5 ?

Who has said it won't work?

Been reading this silently long enough, so why are you trying to shove rpi4 stuff into rpi5? It most probably not work, hell even stuff for rpi5 does not work on rpi 5 as it has 6 or 7 variants by now, version 3 had a long period where a lot of things did not work. They mostly got fixed by now.
You had a very good reply at the start of this topic, install raspios and see if it works, then just add RT kernel and LinuxCNC.
This is comming from someone who just spent 2 days trying to shove ardupilot with autonomous flight into an ESP32!
One part of me wants to say: take the advice and be done with it, the other part is screaming : keep going, make it work no matter how many tell you it wont work!

insmod: error inserting '/usr/realtime-2.6.32-122-rtai/modules/linuxcnc/hm2_pci.ko': -1 No such device

Means the card (5I20?) was not found (no valid cookie read from card)

The two red lights should go out when LinuxCNC starts and loads the
selected firmware into the card.

It may be that some PCI setting was lost when the CMOS battery failed,
a bad PCI/dirty slot. or failed 5I20 card, hard to tell without more data.

If the 5I20 card is bad, the most compatible replacement would be a 5I24
This would need to have similar firmware installed and the card name
in the hal file changed from 5i20 to 5i24.

 

So, while i am waiting for MESA hardware, i have found out that Linear encoders are plugged into Dialog control. 

I have found board: 
Heidenhain 255 180 ZY. I would prefer this one over EXE unit. My glass scales are Heidenhain LS904. I think this is compatible with the Platine?  

Thanks for the reply, Andy.

So if I'm understanding, you're suggesting to use a single HAL PID component per axis but connecting outputs from both of the encoders into that loop?  Interesting idea, I hadn't thought of that.  I have to do some reading up on the PID component and sketch out what the HAL connections would look like.

From linuxcnc.org/docs/html/man/man9/pid.9.html

The gage wheel encoders for the table are not perfect due to their age and design, but I think they should be accurate enough. The motor encoders do have more effective resolution, but that doesn't directly translate to positioning due to where they are located in the mechanical system (i.e., behind timing belts and a ball screw with some backlash).

My long term thinking is replacing the table wheel encoders with magnetic scales, but I want to see how far I can get with these wheel encoders since they're already on the machine and they're already paid for .
 

But... the update of rpi-eeprom-config to version 2712 doesn't actually change the boot order. No longer messages about PI4, so happy its really installed.

Goes through the motions, confirms on screen that the new boot order will be applied at the next reboot. Machine reboots and rpi-eeprom-config shows that the boot order is the same as it was before.
Sigh.

There is an install in the test folder of the clone.
This runs and installs this later version...

Thanks

One thing that might be worth trying would be to use the motor encoder velocity output and the linuxcnc PID velocity feedforward terms (pid.N.command-deriv) and then use the bed encoder for the position parts of the loop.
(This assumes that the bed encoder is accurate, maybe it isn't and _that_ one is intended to be the velocity feedback? )

At attempt to round up some links to other threads on similar topics... some may not be directly relevant.

• www.forum.linuxcnc.org/10-advanced-confi...ectory-planner-et-al
• forum.linuxcnc.org/10-advanced-configura...oops-gain-scheduling
• www.forum.linuxcnc.org/24-hal-components...der-pid-and-backlash
• forum.linuxcnc.org/27-driver-boards/3127...ble-pid-loop?start=0
• forum.linuxcnc.org/10-advanced-configura...-encoder-pid?start=0
• www.reddit.com/r/hobbycnc/comments/118ze...nstead_of_motors_to/

Other related links:

• www.galil.com/learn/online-videos/dual-l...compensation-methods
• granitedevices.com/wiki/Dual-loop_feedback_position_control

sudo python3 -m pip install --break-system-packages pycryptodomex
success.

./rpi-eeprom-config -e run from inside the rpi-eeprom clone directory appears to work, but still fails with RPi4 error.
I can now see on the processing message that it is running an eeprom..2023.bin from outside the clone directory.
To me its odd that there is no install script in the clone and without it, I am stumped.
Thanks again.