Advanced Search

The lathe signals are going to the PLC's I/Os.
Since I'm new to PLCs and have a few mesa card lying around, I thought, linuxcnc was a easy way.
Looks like I need to learn to program a PLC now. If anybody have the firmware for the Arteco PLC (for analog axis control) {Arteco mentions they provide function blocks for analog control or PWM} I shall try to get the existing PLC working or should I just get a new PLC. I have the schematics for the interface between lathe and bar feeder.

One option would be to have SheetCam to not send a material change and then set the material manually before cutting.

 

Quick google shows on eBay a picture of Citizen CM2-PC805-0 it's not a controller it's just a discrete IO card. The interface between your machine controller and the Arteco PLC is done true discrete IO's.
You need the schematic of your machine to be able to wire a new PLC to control the bare feeder. no need for LinuxCNC an overkill for this project.

www.gasparini.com/en/services/assistance/

Hey Scott, sorry to bother you with this, but can you explain this a bit more? Im wondering how we can just do a math to figure it out. Ive been bruteforcing pid stuff for a few days but i cant get my overshoot down to less than 5 steps, but this question comes up on discord a bunch also.

using a 1mm distance for the example to make it easier(?)

ff1 * cmd_d + error * pgain = velcmd

ff1 = number in ini file
cmd_d = old pos - new data? this is the D ?
error = ?
pgain = ini number

edit, holy shit it my issue was a typo this whole time.

my issue is typo related, but the question still stands, is there a math way to figure it out?

IIRC 5 means every 5 cycles

Hello,
I wanted to replace the external Linuxcnc control PC on my mill with a dedicated control inside the electrical cabinet, but I needed something compact that is easy to wire inside an enclosure, which is why I thought of using the RPi5 and a custom designed GPIO breakout board.

 

The board is intended to be used as a 5 axis step/dir platform with 6 extra outputs and 10 inputs. All the Raspberry pins are digitally isolated and the outputs are buffered with 5V @ ~20 mA drive capability. The board has dedicated power connectors for the 5V RPi supply and 24V isolated bus. A 5V step down converter connected to the 24V bus provides the necessary voltage to run the digital isolators and buffer gates.

Here's the new repository with the Kicad design files: github.com/atomicmill/raspberry-pi-5-Linuxcnc-hat

So far I tested the board with LCNC 2.9 and the Probe Basic and PlasmaC UIs. I hope this board is of use to anyone looking for a low cost all in one LinuxCNC control.

I removed the tool change portion of the Gcode and didn't see a plunge to 0, so that narrowed the culprit down.

This is the EMCIO portion of my INI
[EMCIO]
EMCIO = io
CYCLE_TIME = 0.100
TOOL_TABLE = tool.tbl
TOOL_CHANGE_QUILL_UP = 1

Not sure what the quill up means, but there's no position for the tool change to go to, so I guess it defaults to G53.

I didn't find anything for M30.

I suppose the work around is to generate the Gcode and remove the tool change stuff (fusion restricts multiple operations with tool changes for free users) and then add a bit to raise the spindle above the work piece before starting. Because I needed more reasons to dislike fusion.

Thanks for the answer.

Has anyone managed to get DO to work?
I added the following lines inside <pdo idx="1600">, but to no avail.

I'm making chips again . It was difficult to deal with the removal of the chips.

Aluminum is 6082 t6

Here Is a picture of it. 
 

Have you remapped any of your G-codes? (specifically M6)

Other than possibly something hidden in the M6 settings or an M6 remap, I don't see anything that could even move the Z axis at all.

Do you have anything configured in the EMCIO section of your ini file for the tool changer settings? Specifically "TOOL_CHANGE_POSITION = 0 0 0" or anything pertaining to M30?

If you don't have anything configured for the tool change position, (assuming your are using manual tool changes.) Linuxcnc may be assuming a manual tool change position of absolute Z0 (same as issuing a G53Z0 command) for the tool change position. And if you don't have the machine Z0 position configured as the top of Z travel, then I think we have found your problem. Otherwise I have no idea what could be going on.

By increasing the value of refClock to for example 5 ms, what actually happens, do I update the data on the bus every 5 ms or is the data updated every 1 ms of the appTimePeriod anyway?

It's still a work in progress, but the computer is now mounted on the tray where the original controller used to be. It's quite convenient since the tray folds out of the cable cabinet, making the computer easily accessible.
 

Good on you. Nice work getting another Maho going.
Cheers,

Mark

Thank you, Mark. Your MAHO videos are actually the reason we decided to get a MAHO and go ahead with the retrofit.

Aciera,
Thanks to your help, my project is almost finished now. I couldn't have done it without you. First of all, I would like to thank you very much. I am currently planning to put the limits and axis calibrations in the gui, but I do not know where the settings page is, I can change the home page, but I cannot access pages other than the home page.