Advanced Search

I took a picture of the first kinematics on the file, Y axis is interpolation of X an X’
When second kind is active (Y axis is mapped to C and physical Y is not moved)

Hi Spumco!
So yes, at the end of the video link I shared you can see how it looks like, but basically you write the code as if you write it in a plane where the Y axis is mapped to the C axis to a particular diameter (specified with R parameter when calling G7.1)
Example
G0 z0 y0
Y100
Z100
Z0
Y0

That would map the Y axis on the C axis, easy thing (tomorrow I need to machine some parts and I will post the gcode and the finished part)

PD: attached a picture of the new panel working (still missing the label strips)

Hello, 
I assume I'm in the right place, but sorry if not.

I saw a lot of videos like 7i92 and easy-peasy setups and decided to use PC+Mesa instead of smth like SZGH-CNC990(1000)TDb-*.
I have a lathe with the servo (HLTNC HL-T3D driver + motor brake) in Spindle + 2 step motors (for now, but same servos in the future in each axis) + step for ACT for another spindle (milling) + at least 2 optical rulers + encoder for spindle (optional, because I'm going receive position from driver for now and rely on this info, but then going to switch to dedicated spindle encoder).

So I bought 7i95T. And install LinuxCNC and... Pncconf wizard doesn't work with new boards because of PIN naming (as I understand) and this is only the beginning. That's why I need (correct me, if I'm wrong) MesaCT. Am I in the right direction? Or do I have to fix somehow Pncconf (the main problem is that XML connector name is NOT "P1", but "STEP/DIR+ENCODER+I/O" and python to int convert issue)?

Thanks

Additionally. If you are designing a Motion Control Board and want to ensure 110% isolation between the domains, the "go-to" part is the ISO7760DBQR (imho). Tested that all of today. Will endure a lightning, two typhoons, and a russian EMP bomb (granted all that are tiny!!). Others; the part's INs/OUTs are HIGH impedance and so no need to pollute the entire board with pullup resistors. 5V-to-3V3 translation at 25Mhz == PASS. Solid square wave with little to no under/overshoot.

The ISO7760 part was specifically engineered for system automation and motor controls, therefore you can officially ditch those half-useless ancient and medieval 'optocouplers'

e.g. 

Happy Sunday, everyone! 

Got my MPG functional in RIO. Everything works; encoder counts, selection knobs, led, fairly happy about that. Not entirely sure what else I need on that motion control board, probably nothing. And so time to finalize the routing and send out a prototype PCB to test with!! Not that the breadboard set-up doesn't work, it's just a hazardous messland all over my benchtop // flying wires much lol \\

Удалось ли кому-нибудь установить NativeCam на LinuxCNC 2.9.3? Две недели бьюсь с проблемой Python 2.7, а теперь ругается на GTK... Сил нет.

I left the refClock value at -1 , but it gets worse.
The vibration problem is present at every restart of LNXCNC.
So I reset it to 1. It would probably be better to set it to 2 ?

This is a retrofit from a little while ago. It is a massive Italian pipe bender from the '80s.

In principle, it is a simple machine in terms of kinematics, but the size is intimidating. It had dozens of powerful hydraulic actuators and sensors, most of which would destroy themselves if actuated in the wrong sequence.

Two axes were driven by servo motors. We had to keep the old analog servo drivers, which was a requirement. They have resolvers and translators to encoder-like signals in each driver.

The 'bender' axis is hydraulic and uses a proportional valve for control. It is massive, and it’s hard to describe the sheer force this machine has—quite scary.

The kinematics were implemented in G-code, and the final result consists of several O-codes and routines working with custom HAL components to make everything function. In the end, it was about a second slower than the original one.

I had the opportunity to work with a great company that handled everything else. I just had to do the CNC work. It was a three-week job, working an average of 10-12 hours a day. It was worth it. The machine is working flawlessly.



This piece below is one of the most complicated ones in terms of logic. To avoid collisions, the machine has to release the pipe, bend it, hold it once again, and then push it to the last bend. The operator only has to input the dimensions, and the code is able to figure out on its own if there will be a collision and choose a way to avoid it.



This is the typical speed the machine runs

Thanks...I am really looking forward to using this...board should arrive this week...have one issue in the meantime...same pc I had xp on for running mach3...installes linuxcnc with debian 12 from website...something makes it restart and I have to re login..hp compaq with 4.96 mb ram 1.8 ghz...250 plus gb disc space...I do not think it happened when I ran as live boot a few times before installing to disc...

Thanks again!!

The code written by Aciera works, but I'm having problems with permissions right now. I created a common folder between Windows and Linux. I can't give permission to this folder on the lşnux side, even though I used the chmod 777 command.

very good to know

thanks you all

 

Dear Forum,
I started out with the .iso of linuxcnc. Configured my machine with Pncconf Wizzard and realized that I have to compensate for the geometric imperfections of my machine. There are some clues in the forum:

forum.linuxcnc.org/10-advanced-configura...t-axis-errors#105999
forum.linuxcnc.org/10-advanced-configura...onent-and-hal#191290

I'm thinking of writing a custom Kinematic module that takes a Table of commanded Positions vs. measured Positions  and generates (interpolates) the correction and commands the motors (closed loop steppers) to the correct position.

What is the best way to do that?
Do I have to compile the whole LinuxCNC myself?
Is it sufficient to compile the kinematic module? 

with kind regards

Martin

 

rodw wrote:

Refer here. Please restrict this thread to Ethercat installation only
docs.google.com/document/d/1jeV_4VKzVmOI...diY/edit?usp=sharing

No. 59 right ?
link sent back your first post
 

 

You could check if there is 3.3V power, and if not, maybe trace
backwards (3.3V comes from a switching supply that is supplied
by VFIELDB)

 (Check across the larger electrolytic capacitor
 C30 towards the bottom of the card between TB2 and TB3)

But if its not something obvious, the card should be returned
to the vendor for repair/replacement.


 

Looking great, I'm following this project with interest!
Thank you for your efforts!
/Stefan