Advanced Search

This should be fixed with version 008.047. Let me know if it's all good for you after you are able to update, please.

Thanks for bringing this to our attention!

I did a diff on the two readhmid outputs from the re-flashed one vs. another one that I hadn't touched and they were the same for the pins in question. So either the flash didn't take or something else is broken. (there were other differences in the two files)

In any event for now I'm going to go back to making a DB-25 converter since I've exceeded my rabbit hole depth limit

Thank you for sticking with me.

I'm not a expert but to compile this file it needs to link all the #includes so I don't see how that would work without the source code.

In any case doing a RIP installation is not really such a big deal. I work pretty much exclusively with locally built installations, getting all the required build dependencies can be a bit tedious but once you've got past that it's really quite simple.

I suspect the default firmware is a 5ABOB type parallel port firmware
but I can check later.

Hunting on the step/dir outputs is more likely a setup issue than firmware
(since all firmware stepgens are identical)

I can certainly confirm that machine vision can be used with external offsets without any extra hardware (except the camera of course). Be aware though that python may not be fast enough for your application so you may need to use the c++ library of OpenCV (if that is what you are aiming for) and somehow stream the correction offsets to a LinuxCNC real-time component (which will need to be in C).
This may be of interest:
linuxcnc.org/docs/stable/html/man/man9/streamer.9.html

Be advised though that there may not be much help available regarding the actual implementation of your code so you would have to be fairly confident with programming or willing to do a fair bit of your own research. Of course we always appreciate project reports.

and there is no way to just compile this file? just building Linuxcnc from scratch? I'm not very experienced with this, I thought maybe it wouldn't even compile a component, but apparently it's more complicated than I imagined it would be.

AxisGUI has a simple interface that lets you select the axis and then you can use the '+' and '-' buttons to jog.

Ah, so the .pin file is the descriptor for the .bin file. That's helpful

I have 3 of these things and this is the first one I've ever flashed. Presumably the factory image has some standard pinout that miraculously matched the pin in's of the other two devices. That's what I want to get back to.

I put the thing on the scope and it's just not right. If I scope the original computer with native parallel i get a nice clean step signal, but if I scope this card I only get intermittent step pulses. But the width of the high pulse is the same as the native parallel.

So I want to get back to the native firmware just to see if the intermittent problem goes away. I think it will because when I first go this card it did work on a different machine.

Thanks for the explanation on the .pin and .bit files - that helped a lot.  Do I need to worry about Table1 and Table2 mapping?  Or is that all taken care of in PNCConf?
 

Ok...I will look that up on how to...can all 9 axis be shown on jog panel as well?

Thanks!

You were correct about the 'ON_ABORT_COMMAND' not working with all those commands. I will go the subroutine route.

I realized I forgot to comment on the MDI_COMMANDS section. Those are formatted that way to show up as buttons in the UI. Everything after the comma is treated as the button text. I just used an example bit of code from the pull request on github to set it up. github.com/kcjengr/probe_basic/pull/84

That example code uses the INI format from QTDragon.

Not sure if you can have a mix of G- and M-commands like that for an 'ON_ABORT_COMMAND' so try commenting out this line in your ini and see if the error still occurs:
If that fixes it then you may want to use a subroutine call instead. Something like this:
and create a subroutine 'on_abort.ngc' with all the command you want to run on abort

Actually, let me retract a bit of that -- I've tested three models:

- ECT60
- ECT60x2
- ECR60x2

I haven't tried any of the other ECR* models. IIRC the code for the ECR60 and ECT60 are basically identical (see github.com/linuxcnc-ethercat/linuxcnc-et...ces/lcec_rtec.c#L129); the only real difference is that one shows up in `lcec_configgen` with a modparam for `motorResolution_pulses` while the other has `controlMode`. Under the hood they're treated identically. The ECT60x2 and ECR60x2 seem identical, and IIRC the specs for the normal ECT and ECR models looked the same with very minor differences.

Can you also send me `ethercat sdos` for your ECR60? That'll help. Filtering it down to juse `ethercat sdos -p XXX`, where XXX is your ECR's Ethercat slave ID would be best.

You just need to select firmware that matches your breakout.
The original firmware not really relevant, its just there to allow
access to the card for re-flashing to the desired firmware.

.pin files are just text file that list the included firmware modules
and the pinout of a specific firmware. These are created with
mesaflash and can be used to check firmware compatibility
with a given breakout.

.bit files are firmware files for Xilinx based FPGA cards

.bin files are firmware files for Efinix based FPGA cards

Hello,

Thank you so much for all the help. I just wanted to confirm that applying external offsets through a custom HAL with the data obtained from a visual tracking algorithm doesn't require any additional hardware, i.e., external controllers. The only reasons I want to be 100% sure is that I currently run my cnc using mach4 and it has been made clear to me that mach4 would require an additional controller to perform what I am looking to achieve. I reached out to a motion controller company and the cost for customizing their motion controllers was very expensive. So, before I transition to LinuxCNC and overcoming the learning curve, I wanted to confirm with you if my project could all be achieved purely with software on linux? Thanks

If you want to modify the source code you would need to clone the source code and build linuxcnc locally on your machine (called a 'Run In Place' or 'RIP' installation):
linuxcnc.org/docs/master/html/code/building-linuxcnc.html