Advanced Search

Studio_TC wrote:

Startup/Boot Mode = Legacy Only (options are "Auto" or "UEFI Only")

Try UEFI only, it works on some PC's just fine.

Hi,
Just received a refurbished Lenovo ThinkCentre M710q and am working on installing the latest "LinuxCNC 2.9.4 Debian 12 Bookworm PREEMPT-RT ISO".  Got it flashed to USB via balenaEtcher and can run the installer and the live version.  With the installer, I make it to the partitioning section and it fails with:  "The ext4 filesystem creation in partition #1 of SCSI1 (0,0,0) (sda) failed."  I'm hoping someone knows the magic sauce to get this machine running...

I have been using the default guided steps (Guided - use entire disk) for partitioning, not trying to do anything fancy.

In the BIOS I have set the following:
Security/Secure Boot = disabled
Startup/CMS = enabled
Startup/Boot Mode = Legacy Only (options are "Auto" or "UEFI Only")

I also tried disabling all the "Intel(R) Manageability", "Intel(R) Software Guard Extensions", "Intel(R) SIPP Support" in the Advanced section.  As an aside, should I disable "Turbo Mode" in the "Advanced/CPU Setup" section?

In the live session I can get to "GParted", should I try partitioning the drive there before running the installer or is this a bad idea?  I was hesitant to try that...

I'm not sure how to proceed, any advice is welcome.
Thanks
Tom

PS/FWIW I recently installed LinuxCNC to a laptop using ISO/balenaEtcher with no problems (other than the laptop not being suitable).

Hello Kieran,
A hardware driver such as mesa or ethercat when run will expose hal pins. You can use halshow to see the driver pins. These driver pins can now connect to hal components pins.

linuxcnc.org/docs/html/hal/tools.html
 

Not sure how the potentiometer was used but It could replace home/limit switches
(and since it's absolute no homing cycle would be needed at startup)

The encoder is possibly used for both position and velocity feedback

Is there another motion axis?

 

The machine has been incredible for 10 years. I think something got messed up with the files. I sent them to the guy that programmed the machine to see if he could figure it out. Thank you.

Does the pot have hard stops?
How many turns does it take to hit those stops, if any?
How many degrees can it turn, if it does less than 1 turn?

Kieran wrote:

Im thinking I can send-cut-send a panel, with switches as desired, and use a keyboard USB "brain" to read them. then basically map them however needed at that point. that's the easier part for me. Learning the computer/software GUI stuff is gonna be a learning curve.
 

S-C-S is a great idea.  And I sympathize with you about the software learning curve... 

OK, I understand.

But what is the operating condition of this potentiometer?

- Is it precise enough for me to use it for position control and the encoder for speed control?

- it only allows the machine to be homed without moving to a homing switch (the embroidery machine doesn't have a home and limit switch).

- does it allow homing and limit switches? If so, would a potentiometer for travel limits be useful in addition to the soft limit, or is the soft limit enough?

 

@Pippin88,
Why the report?

It looks like the G code is trying to reference / point to another file?

How was the g code generated?

Am I understanding correctly...   Linuxcnc "talks" to Hal, Hal used XML to translate to the ethercats. so xml is like an " hardware even more abstract layer". So I should assign "X Y Z A B" names in the XML and then Hal can control the servos from then on by those names? Im assuming the ethercat slaves show up in the order they are physically plugged into the chain? or is how they are plugged together not matter much? I guess I have to dive into the "debug software" for the servos too and set up the electronic gear ratios and whatnot. 

Thx a lot!!
I will take a close look at both. Finally a first point of reference from which I can work my way forward.

Then let's keep each other up to date.

I think I can see a way to do this, as Andy explained I will take a look at this when I get a bit of free time.

Its a bit tricky because the first index detections ( used for latching the position ) will zero the position
so the latched count + encoder position must be used for feedback until the final homing to index
is done.

What I was thinking is that the detection of the desired position difference enables passing the index detection
to LinuxCNCs motion component.  

Nice welds on your control panel.
Mark