Advanced Search

I'm just thinking that a XYZ jerk limit smoothing 'postprocess' app would enable people to test the algorithm.
Let's put this is on the list. We could output a gcode file with a certain mm resolution. This is the easyest way.
I will not split a g2 in multiple G2's with different a,b,c values.

Ok i just tested when there is only a,b,c move with no xyz. This now works.
When there are xyz values it will interpolate the abc over it. But with tiny xyz values and huge abc values, there i have not
yet a solution for in the code.




 

How do I speed up the homing function on each axis?

1. increase HOME_SEARCH_VEL
2. increase HOME_LATCH_VEL
3. think about homing two axes (eg X and Y) at the same time (by assigning the same number)

www.facebook.com/share/LVbRyNUEmDktgdfu/?mibextid=A7sQZp

Or

matrix.to/#/#qtpyvcp:matrix.org

There is really only the way I linked a few posts back for the correct user installation.  Maybe if you provide all of the information regarding your pc, iso used,  version of linuxcnc etc.  But really if you follow the linked instructions verbatim it just works unless you are doing something unique?

kcjengr.github.io/probe_basic/development_quick_start_apt.html

Follow these instructions and it will work.  I do it every few days helping others and it has no issues. 

If you have other installation attempts,  you may need to start with a clean Debian install. 

We would like to have it packaged with linuxcnc,  but unfortunately it was rejected because of politics.  As time and changes happen we have had to change the installation method.  The current will hopefully be the last time that has to happen.  Message us on the matrix chat for faster assistance,  or join the Facebook probe basic group and join the chat there. 

Chris

 

I am so frustrated with this,I have installed and reinstalled about 10 times now and it doesnt even show up in the Linuxcnc menu.
i have tried following the youtube videos but they are 2 years old and the installation has changed since then.
Could someone possibly put it in an iso with linuxcnc that would work? I will pay for it, i have spent about 6 hours with no success.

Are you Chinese? If yes, I can guide you step by step through Sunflower remote control. My QQ is 304394405

riogui is a pretty good way to config the board modules, and it also spits out a pretty good working config. Like all of the config utilities there is an amount of hand config editing required.

Checkout the this github page, shows the riogui.
github.com/multigcs/riocore

Hey I'm so happy rite now I just have to show my appreciation for what I've got!  First off I want to thank PCW for being a genius and creating the Mesa board.  I also want to thank TommyLight for the inspiration to make a goal of fixing my machine.  It originally had a spindle issue and sucked at properly functioning as a PMW based VFD.  So replaced it with a T6 from StepperOnline threw it in the case and slapped a Mesa 7i96 to in and it fired rite up with no issues what so ever!  Anyways I'm so proud of this and want to let everyone at LinuxCNC know that I really appreciate all this!  

If there's anything I can do to help or contribute to the awesome CNC movement let me know.


Lol, a quick question.  How do I speed up the homing function on each axis?  Thanks




 

You can use pncconfs discover option to more closely match the firmware.
(then disable the unused P2 (7I75) functions by setting the number of
encoders to 4, stepgens to 4, and pwmgens to 0)

To get the correspondence from say P3 stepgen 0 to 7I85S TB pin
you can run mesaflash with the --dbname option:

sudo mesaflash --device 5i25 --dbname1 7i85s --readhmid

Alright, I got the card flashed with the 5i25_7i85s_7i78 firmware. (I think, info on the card printed by Mesaflash is in the text file attached.)
Now, can I add the firmware to PncConf so I can use the nice dropdown menus for at least most of the configuration or will I have to go straight to the hal/ini files?
And then, how do I figure out how the pin numbers on the 6i25 correspond to the pins on the daugher cards? I guess the general I/O taken care by the 7i75 breakout board will be easy, but what about the 7i85s?
Say, I have the step pin of the x axis driver connected to pin 12 on the 7i85s TB2 connector, and its function is /TX0. So I check TX0 from the connector pinout in the 7i85s manual, and find out it corresponds to pin5, IO8 on FPGA PRIM I/O and IO25 on FPGA SEC I/O, so is my I/O number for X Axis stepgen 8, since I have the boards connected via the primary DB25 connector?

Thanks Chris.

Rather odd but, then I do have debug running so expected.

Appreciate the feedback.

You really should give Ethercat a try!

It's not for lack of interest. Seems to be the way things are headed but I'm usually in need of all kinds of different digital/analog IO, encoder inputs and generally more unconventional setups. When you factor in support for such things MESA hardware is just really hard to beat.

That board looks sweet!

If you have analog ports that could be broken out I would buy one. Have a few projects in mind that need a analog port (thinking a height control for a CNC CO2 laser that follows the warped wood..yes a thcad would work but a few more projects need 10+ analog input). 4-20ma is preferred over 0-10v due to noise issues.


I like the plug in design for stepper motors as can probably find cables so we don't have to wire them by hand.

Personally prefer Ethernet cables for everything as they are much easier/faster to crimp by hand than those white connectors and I can wire directly to a machine 20 foot from my control panel. Wire and connectors are cheap and abundant

What makes projects like this succeed and become interesting is good setup instructions and documentation. Bonus points if you can make a pnf config style config generator. Time is much more expensive than the boards themselves, I'll pick a $250 mesa board over an unknown one with no documentation because it's going to save me $1k+ in time.

Hope you keep developing it. Looks awesome!

I'm just thinking that a XYZ jerk limit smoothing 'postprocess' app would enable people to test the algorithm.

But i heard, when a,b,c are involved it falls back to no look ahead mode.

Yes that is true, it does not make much sense for multiaxis. Would only be of interest for XYZ axes.
Getting advanced look ahead/ velocity blending for multiaxis working is pretty much a prerequisite for multiaxis operation.