Advanced Search

Great, thank you for the info. Can I use solidworks and get the file into Inkscape from there? (Save as a 2D dxf file and import it into Inkscape)

Also is LinuxCNC free?

I have more questions, in the case of only using limit3 and pid+stepgen how could I establish the limit switches for each axis? This is within the motion control if I'm not mistaken, well I hope for fresh ideas hehe, greetings

Why 2 of 7i92? Why 7i80?
I used a single 7i92, 7i74, several 7i70 and 7i71, and 7i77.
Since there is no easy to find 7i77, i would use 7i92TM plugged directly into a 7i85, then a 7i83 for the analog outputs, and add 7i70 inputs and 7i71 outputs as required.
All of those can be easily found, well if there is no 7i92, get a 7i96S, adds another sserial and some inputs and outputs, and some stepgens you probably have no use for, but will come handy if you ever have to change drives, new ones can do step/dir, most of them.

I have several foam cutting machines in daily use for many, many years, controlled by LinuxCNC, so if you are asking
Can LinuxCNC run a 2 axis machine?
Yes it can, from 1 to 9, each can have several joints, and much more if not assigned as axis.
And yes it can run a foam cutting machine, be it 2 or 3 or 4 or 5 axis.
For CAM use Inkscape, it is free and can output gcode directly usable by LinuxCNC. It can not do 4 axis, but does 2 axis easily.

Update of the Hurco retrofiting.
Did run it today with  the old controller. It i runs but have issues with UI freezes up and going in to sleep in middle of programming.
So the new controller solution is needed before could use it to any serious jobs.

Regarding this ordered the mesa boards the other day but EU-Surplus did not have the 7i80 in stock, neither does Mesa them selves.

So the question is now should i wait for a 7i80 card or would it be possible to issue the like of two 7i92 instead?
I have read somewhere here on the forum that the real time UDP controller cards should be directly connected to the PC and not through any switch.
So with a setup of two 7i92 instead of a 7i80 then I need to use a switch in between. Would this work and is there risk to use such a setup instead?

I really have no real urgency in getting the machine ready so waiting a couple of months is okay as it is right now.

Hello, I am looking for controller software that will run a 2 axis (horizontal and vertical axes) foam cutting. I see forums like  these  discussing this 2 axis software issue, but no one comes to a concrete solution.
This machine is meant to cut airfoils for formula racecars, so no need for 4 axes.
https://lh7-us.googleusercontent.com/v2mEQDZu6rNEEwdOJjn_bqSS542OCyz9Hb--cdkY-NKJMqVhxI6ZeEZjKNL3_-AGZNIkmkwFVAzFDNMzs8zBoPqbQ6XiDlCgkUH5jxiVCtE_odPuoLfngF4SgqBucfGOXqd4j5oGwJjjcRGlOD77W4bZWQ=s2048

I managed to get the drive to OP state when starting LCNC, but only when I uncomment all PDO adresses in the xml.

Its done in the hal file when motmod is loaded:

loadrt [EMCMOT]EMCMOT servo_period_nsec=[EMCMOT]SERVO_PERIOD num_joints=[KINS]JOINTS

It's done this way typically with ini file macro substitutions but can be done directly in the hal file without
any reference to the ini file at all...


 

No double posting.
The other topic deleted.

Try adding this function:

apt-get update
apt-cache search linux-image

shows a newer kernel:  linux-image-6.1.0-20-rt-amd64  PREEMPT_RT

Hmm. I wonder... would that be compatible with the stock linux.292 distro ? Should I try the kernel update? Feeling luck punk?

Funny thing... the networking appears to work fine (but sure, browsing and downloading is perhaps not as time critical as CNC control!)


Update: Went back and checked ping with the older PC and the same MESA card.... results are consistent around ~ 0.4ms.  That seems much more like it, and seems to further confirm the MESA card is behaving correctly.
 

Reported kernel is: 6.1.0-18-rt-amd64
 

Thanks for the pointer at least I now know what to look at. I tried manually uploading the config file but tftp timeouts (ascii & binary modes) as well:

tftp> status
Connected to 10.10.10.10.
Mode: netascii Verbose: on Tracing: on
Rexmt-interval: 5 seconds, Max-timeout: 25 seconds
tftp> put /tmp/config.txt
putting config.txt to 10.10.10.10:/tmp/config.txt [netascii]
sent WRQ <file=/tmp/config.txt, mode=netascii>
sent WRQ <file=/tmp/config.txt, mode=netascii>
sent WRQ <file=/tmp/config.txt, mode=netascii>
sent WRQ <file=/tmp/config.txt, mode=netascii>
sent WRQ <file=/tmp/config.txt, mode=netascii>
Transfer timed out.

I will download the remora source plus the build system to add some debug messages and  hopefully identify what is happening.

The latency test only checks dispatch latency, it does no I/O.

The long and consistent ping times will make operation impossible
(at least at a 1 KHz servo thread rate) Not sure what is causing that,
not something Ive seen before.

The CPU speed issue includes the latency histogram, it doesn't display
there either, as expected:

 


I wonder if the distributed kernel is too old for the N100?

I'd like to write about my experience with the unspeakable Asus PN 50 E1 barebone PC with AMD graphics, for those who are interested.So in 2021 I made the same mistake for the second time in25 years by buying another PC with AMD graphics. The second time, too, thermal problems occurred sooner or later.I can't say whether the issue with OpenGL is causing the thematic problems, but with a little research on YouTube you can find videos that deal with similar problems to mine.The fact is, however, that the Mini PC has terrible cooling. Well, how can it be well cooled when everything is so small?There is indeed very, very little space to get anywhere near decent cooling. In my opinion, the ASUS engineers didn't make much effort to cool the CPU even remotely well. In my opinion, the whole cooling concept looks like planned obsolescence.I disassembled the PC and studied the cooling concept. I doubt that this small radial fan can really draw cool air into the case from the outside. If you only know a little about aerodynamics and/or ventilation, you realize that the fan can't possibly draw in air efficiently through all these slots and air gaps in order to dissipate the heat from the processor in the cooling element.I have now removed the grilles and nets from the housing openings for the intake and exhaust.The transition from the CPU to the cooling element did have heat-conducting paste. The question is of what quality was it? How good is the cooling element?The intake side of the radial fan is just 4mm above the components soldered on the motherboard. If you know that a fan should have a trumpet-shaped intake side and see here that there are only obstacles and sharp edges at the intake, you know or suspect that no air is transported here but only circulated and gets hotter and hotter, which is the reason that the CPU temperature rises from 53 to 88° within 15-30 seconds. Crazy, isn't it?Now I will have to make do with the modified inlet and outlet openings and newly applied heat-conducting paste. That will have to be enough to keep the thermals constant, otherwise I'll have to get a new PC.