Advanced Search

Thank you Collins.

Attached video shows the 3 interpreters tools moving around.

There is no motion component or whatever. The tool is just jumping to the consumed gcode position.

One of the problem's is when adding gcode segments to a infinite program, it has to check for duplicates,
and there must be a limit to the gcode preview. Otherwise we get a buffer overflow.

This run's quite stable, fixed some runtime bug's.

Was looking for info about raspberry, do you suggest to make a new buil on pi 4 because it's more stable? Or it's worth go with the 5?

you can use this Spartan6 module, but have to setup/modify 'riocore/boards/Spartan6/board.json' and change package and type for your needs.

programming will not work over w5500 you need to write the bitfile into the flash.

---

@all
WINDOWS-WARNING:
rio was not intended to run on windows, at the moment i can't give windows/mac support for it.
I just made the system a bit more platform independent.

i normally don't use windows at all, but i think if you customise the generated makefiles a bit, you could generate the bitfiles there too, but very experimental .

Another question for Meister:

FPGA Module 2.00: Spartan 6 FPGA Board.

I would like to use that FPGA board. It has SPI / JTAG connectors for programming EEPROMs. Is it suitable?

I assume that the W5500 uses another SPI module inside the FPGA after programming for the data communicaktion with LinuxCNC. Therefore, I think it the configuaration EEPROMS cannot be flashed via the W5500. Is it correct?

Does the RIO System write the FPGA everytime when getting online or is the data permanently flashed on the EEPROMs?
 

i think you are on the wrong branch:

git checkout dev


BTW: very experimental, only for the gui and to generate the config-files,
i think, building bitfiles will not work yet on windows


 

Hello community,
I have another problem to solve, namely my feed potentiometer only works at mm/min.
It has no function at mm/rev. The potentiometer for rapid traverse always works.
Is there no other way or can it be influenced?
If you have to run in a program, a potentiometer like this is very helpful

Thanks and best regards
Olli

Hi Meister,

Unfortunately, I cannot make it run under Windows.

I have changed:

file_path = r".\riocore\__init__.py"
if os.path.isfile(file_path):
sys.path.insert(0, os.path.abspath("."))

When running rio_setup, the first problem, I cannot import halpins from riocore.
Directory is OK and the __init__.py file is there.

Perhaps you can help me out. Thanks a lot!

Thank you. I will study more.

And for LAN9252, it is better to copy the whole lib/soes from Easercat 2000 and go from there. That should work right away.
 

You can also add a PDO just for debugging, I had several of those during development.
Just to assign Obj.D1 = interesting_value and watch that value.
An alternative to the index shift is to have a counter counting up when the callback happens, there is
actually one already in my code "encCnt" which increments every time the index callback is called.
Just quickly checked in the debugger, here the callback is triggered when index passes by.

I understand the speed, it's hard to see the index pulse.
What I don't understand is when an index is detected, how does your encoder command element change the state?
For example:
Obj.IndexStatus = Encoder1.indexHappened();
Checks if an index pulse has occurred (uses the indexPulseFired variable).
If it has, it returns 1 and resets the state.

Obj.IndexByte = Encoder1.getIndexState();
Reads the digital value of the index pin directly (0 or 1).
Used to check the current state of the index signal.

This information, even monitoring with Twincat, does not detect any.

I am now starting to practice debugging and serial monitoring.

And next I think I will try EaserCAT-2000 with AX58100, but it is still difficult to change from LAN9295 to AX58100 with the original firmware of EaserCAT-2000.

The Index signal is short, in EtherCAT it is only one servo cycle long so you will most likely miss it, or the linuxcnc tools will miss it. Unless you happen to position the encoder exactly where the index signal is, then you can see it on the IndexByte.
Use halscope to watch the index signal in linuxcnc. Set sample time short enough to capture all data.