Advanced Search

I can't duplicate this with 2.9.8 or master (2.10-pre1)
sounds more like a hardware issue of some kind.
 
 I guess a software issue is possible but this is the first time I have heard of this.

There were some sserial driver changes (bug fixes)  but nothing specifically
in the sserial encoder section, and there have been no encoder related
firmware changes to the 7I73.

You might check if the encoder still counts when both flat cables are disconnected from the 7I73

Do the other 7I73 MPG encoders have the same error?

Hello Tommylight, thank for your time and help !
After a good night, the ping 10.10.10.10 finally works (idk why), i managed to setup the pncconf wizard and open linuxcnc, but still a connexion issue i guess because i get several warnings at startup :
"Unexpected realtime delay on task 0 with period 1000000."
hm2/hm2_7i96s.0: error finishing read! iter=1598"
Moreover, my homing switch does light up the mesa board and flicker when i use them, but they appear as fixed in Halshow (show on but no flickering). I suppose the connexion break at startup of linuxcnc ? I can't figure out why and it so frustrating !

andypugh wrote:

cnbbom wrote:

would that be it?1 drive.google.com/file/d/1BwlJet6ZcWpXvAUzD2YcM9vvnPTuclaX/view   2 drive.google.com/file/d/1xICFF-2mq_rz-vi.../view?usp=drive_link 3 drive.google.com/file/d/14Z10IOZI5bvgmf_.../view?usp=drive_link 4 drive.google.com/file/d/1fwGb5Qo9pw8lGLp.../view?usp=drive_link


Possibly, though I would need to apply for access as I can't currently view them. 
 

Is this what you were looking for? Were you able to download it after I checked the Google sharing option?

spiking is present maily during movement at same vector .. when vector of speed is different, spikes are gone

Hi,

I am using LinuxCNC for years now and my Hardware is since 2020 stable. My hardware uses a Mesa 5i25 for the main job and an 7i73 as remote with an handwheel. 

After upgrading to 2.9.7 ( with PC hdd and OS update) I found that my handwheel is always counting, even when not used. I have the 7i73 on SmartSerial port0 loaded with:
hm2_5i25.0.7i73.0.0.enc0.count -> is incrementing at max speed right after startup. So I double checked all Hardware even with an oszi, but the electrical signals are OK.

So I tested this code in hal, end replaced the HW Counter with an SW Counter. Then all is working as expected...
Even when it works now, I would like to understand the reason behind to check if the is the even more trouble, which I didn't find till now.

Question:
Where there relevant changes in hm2? My old LinuxCNC before was 2.8 ....
Is a "Firmware" Update neccesary ?
Could this be an defect inside the 7i73? Is something similar known?





 

the most easiest way from my point of view at everything it ... configure TC3 + you hardware and then configure the lcec itself...

if something it not listed you have to know which type of encoder feedbeck it is and then choose right one from DriveManagerX ... then setup the axis in manual jog and everything 

at resolver you had have to know which type, frequency and poles it is ... that all then grab these data and configure the in the Drive manager

my own driver with integrated homing subroutine which is working same as at CoE drivers is here

(TC3 machine with win10)configuration tool which grab export TC3 master and transfer everything to lcec master shape is here

after succesfull configuration of TC3 just export and copy paste to lcnc folder ... its more easy than ever 

lowest scan time which I get from AX5YZZ is 250us scan which I am currently running same as at shared example

I don't know what the problem is. It's always the left Y-axis that sporadically fails to start during homing. I always have to turn the machine off, and then it usually works. There haven't been any problems after homing so far. I can rule out wiring issues because I would get an error message.
I think that for some reason, the servo occasionally doesn't receive a start command for homing and therefore doesn't start.
The error must be somewhere in my configuration. 
 

Great! Thank you

Hello everyone, after a long break, I've started modifying my Hruco again.

A quick overview of the setup: I have a 7I97 mainboard, and I've connected a 7I74 to it to expand the ports.
I've connected a 7IAO (Port 1) and a 7I89 (Port 2) to the 7I74.
I noticed that the 7IAO also has a Xilinx Spartan-6 FPGA, but I assume it already has a basic bitfile on it that I could flash via the pins on the chip, but I don't have to if it's not necessary.
Am I correct in this assumption, or do I also need to flash my own bitfile onto the card?

Now for the second part of my question.
I wanted to create a suitable bitfile for the four cards I mentioned above.
What software would be best to use for this?
Since the FPGAs are from Xilinx, I'll probably have to use their software so the IC can understand the bit file, so Vivado, right?
I also saw something online from Efinix, but those are different ICs.
The software probably won't be compatible with them, will it?

yourmachine.ini  file,  in  [TRAJ]  section  POSITION_FILE = posfile.txt   stores machine positions at shutdown. If this is not set, the machine starts with all axes set to zero

 

Can you elaborate on what are these architectural issues?

andrax wrote:

Have you ever had the problem where one of the Y-axes doesn't start moving during homing and the gantry is crooked?
That happens to me a lot. I have to restart the machine and then it works again.

That should not happen, ever, and it did not happen as i have probably over 15 machines with tandem joints, some industrial ones with over 400KG heavy gantries.
You should check the wiring first, then test the failing side drive.
Are you using the same drives mentioned here?

I've checked all the reported spikes, and the issue is not the new s-curve in itself, but the trajectory planner architecture, we will have to do a much more involved refactoring, will look at Tormach work, maybe talk with Rob if that happens, and take it from there.
The current scurve is best that can be done within the limitations of the current tp architecture.

Thanks for the reply
I tried corexykins, didn´t work because i use uspace?
so i went with KINEMATICS = corexy, which then caused problems with loadrt [KINS]KINEMATICS and loadrt [EMCMOT]EMCMOT...
Why are home switches necessary? I dont have the machine connected at the moment since I´m only trying to get the software to start.

Is this combination of software even the right choice?
I´m building a small portable plasma cutter while using some parts of an old 3d printer and i would really like to use corexy.
Are there other GUIs that work better with this?

...

Edit: Launched linuxcnc on terminal, getting "bad spi payload"
(Bad SPI payload = 32393a38)

The value of the payload is strange; it's almost half value of what you are suppose to get:

#define PRU_DATA            0x64617461     // "data" SPI payload
#define PRU_READ              0x72656164  // "read" SPI payload
#define PRU_WRITE             0x77726974  // "writ" SPI payload
#define PRU_ESTOP           0x65737470  // "estp" SPI payload

I'd try with a lower SPI frequency.