Advanced Search

Thanks! I will check this out.

hey there,

i have this problem, that status time is not counting, 
im using fusion 360, it has been counting before

idk if it´s fusion or qtdragon

hope for some advise

merry christmas every one !
 

need more information on your build, what debian version what kernel version etc... the sim here is working as expected so i would assume the issue is a compatibility problem you are experiencing perhaps or maybe the custom_config.yml file needs to be updated.

have you tested using the simulation version of probe basic?

compare the sim files that are updated in the config folder to the current machine custom_config.yml file

If you figure out which input does not work, if you have a voltmeter or just an LED
you can trace the signal at the 7I76's  DB25 connector to see if the issue is with the
7I76 or 7I92.
 

Have a look at the 'message' component:
linuxcnc.org/docs/2.9/html/man/man9/message.9.html

Suppose I have a python script running that will detect some sort of error condition, how can I get it to display a message in axis, much like the typical error messages displayed in the bottom right  for example "Can't do G1 without a feedrate"

If I need to do this from HAL, I should be able to get the Python script to activate a HAL pin.

Thanks,

If I plug the complete connector, to which the encoder is connected, from the 7i76 into the 7i85, then it works.
So it can't be the cable or plug.
Possibly the input of the 7i76 is not OK.

Glad you got it working. That is strange that the product code would be different than what the drive software shows. To me that definitely seems like a bug in the drive firmware or setup software. Looking at the ESI file from Leadshine they only show 0x10c2 - 0x10c6 as product codes for the EL8.

OOOO i am pinning this!

royka wrote:

That's not the mesa I meant, but try otherwise this image with Linux 6.12 rt:
drive.google.com/drive/folders/1BXbeBIY8...7N4Kli9-?usp=sharing

i'll try that image on opi4 non-lts , nice it got 6.12

Royka , my armbian from minimal to cinnamon 25.02 trixie + opi5 + your rt 6.1.75 run linuxcnc with mesa 7i92 like a charm...better latency then my i5 2500 .

Thanks for all . 

 

If A does not change, that sounds like there is a problem in the "A" encoder signal path
This might be:

1. Encoder connections wrong.

2. Bad 7I76 encoder input

3. Bad / loose DB25 cable

4 7I92 bad/damaged input

Also which pin does not change? A or B?

earlier post:

 When the X-axis moves, the yellow LED at A changes constantly between on and off, B remains on, Count changes between 0 and -1

latest post:

here “Count” only changes between 0 and -1, “B” flashes when I turn the encoder and “A” lights up constantly in the HAL show


 

 I connected normal input on IO 34 qnd 35, it works fine. But how should I tell the system the stepgen index for axis X is on the 36 and for the axis z on the 37 ?

You cannot, the index mapping is fixed by the firmware. That is, GPIO 34 is always the index for stepgen 00,
GPIO 35 is the index for stepgen 01 etc. You can, however map any stepgen to any axis.

So, ich habe mir mal einen normalen Drehencoder besorgt und den an die 7i76 angeschlossen. Dort wechselt "Count" nur zwischen 0 und -1, "B" blinkt wenn ich den Encoder drehe und "A" leuchtet konstant in der HAL Show. "Velocity" und "Raw Counts" tut sich gar nicht. Stecke ich nun um auf die 7i85 (Ausgang 02), dann funktioniert alles wie es soll, "RAW Counts" und "Counts" werden gezählt, ein Wechsel zwischen "A" und "B", also alles so wie es sein soll. Keine Ahnung, warum das so ist. Jetzt warte ich noch auf 2 GT2 Riemenräder und den Zahnriemen, dann kann ich das Ganze mit Motor testen, ob es dann mit dem Gewinde funktioniert.

Gruß
Olli

---

So, I got myself a normal rotary encoder and connected it to the 7i76. There “Count” only changes between 0 and -1, “B” flashes when I turn the encoder and “A” lights up constantly in the HAL show. “Velocity” and ‘Raw Counts’ do not change at all. If I now switch to the 7i85 (output 02), everything works as it should, “RAW Counts” and “Counts” are counted, a change between “A” and “B”, so everything is as it should be. No idea why this is the case. Now I'm waiting for 2 GT2 pulleys and the timing belt, then I can test the whole thing with the motor to see if it works with the thread.

Greetings
Olli

Mint 22 with LinuxCNC 2.10, sim works.