Advanced Search

Well, the only person who knows for sure is the release manager.
And he thinks that a 2.9.7 is very likely given that one of the fixes in 2.9.6 broke the WHB... pendant driver.

But he is also looking at an ISO based on Trixie, but that will need a new archive signing key as the current one is considered too short.

Please submit your work as pull requests to the "official" ISO building repository rather than going off on your own with this.

The question is whether the drives can do preset velocity with no commutation.

ChatGPT (or any AI) is going to do very poorly with anything that hasn't been done before.

Some of the drones i have built, and more in the works, mostly 3D printed

langdons wrote:

Certainly seems that way.

Aciera literally stated there would be another release, did he not?

But when?

Not helpful, I was looking for a definite answer. Not the words of someone whose words soothe them.

Try:
min_limit = -250
max_limit = 0
home = 0
home_offset = 0 or -3 due to not stopping at the switch
Notice there are two instances of min/max limits to change.

Perhaps you will find this post helpful: forum.linuxcnc.org/49-basic-configuratio...iguration-and-homing
It's really old though.


I recommend you make a new config from scratch, if you use either StepConf or PNCConf, and play around with the limit switch settings so moving past the home switch is "allowed" up to a certain distance.

Bear in mind that the default values are default for a reason.

If your new config does not work, it will probably be easier to scrap it and create a new one.

Keep your current config as a fallback option though.

Certainly seems that way.

Aciera literally stated there would be another release, did he not?

But when?

Set it only to HOME, not HOME+LIMIT, that way you can technically cross the switch if the machine is not homed.

In fact, you can probably allow movement past the home switch properly in the config or stepconf or whatever.

I don't think the home switch needs to necessarily be at the end of the travel.

Mid-travel home switches should be possible.

Referencing this post forum.linuxcnc.org/29-forum-announcement...-9-6-released#336514 is there a possibility of a 2.9.7 release or something along those lines.
The reason I ask as I have been working on trying to create an ISO that doesn't suffer from the "raspi-firmware" issue. And I would like to have a dialogue with regards to releasing a 2.9.x ISO.

Yes, that’s the weird part. It is correct.




INI for Z it’s like this:



[AXIS_Z]
# MAX_VEL & MAX_ACC need to be twice the corresponding joint value
MAX_VELOCITY = 66.66666666666667
MAX_ACCELERATION = 1500.0
OFFSET_AV_RATIO = 0.5
MIN_LIMIT = -0.0
MAX_LIMIT = 250.0

[JOINT_2]
TYPE = LINEAR
HOME = 250.0
FERROR = 10.0
MIN_FERROR = 1.0
MAX_VELOCITY = 33.333333333333336
MAX_ACCELERATION = 750.0
# The values below should be 25% larger than MAX_VELOCITY and MAX_ACCELERATION
# If using BACKLASH compensation STEPGEN_MAXACCEL should be 100% larger.
STEPGEN_MAXVEL = 41.67
STEPGEN_MAXACCEL = 937.50
P = 1000.0
I = 0.0
D = 0.0
FF0 = 0.0
FF1 = 1.0
FF2 = 0.0
BIAS = 0.0
DEADBAND = 0.0
MAX_OUTPUT = 0.0
# these are in nanoseconds
DIRSETUP = 10000
DIRHOLD = 10000
STEPLEN = 3000
STEPSPACE = 3000
STEP_SCALE = -1000.0
MIN_LIMIT = -0.0
MAX_LIMIT = 250.0
HOME_OFFSET = 260.000000
HOME_SEARCH_VEL = 16.666667
HOME_LATCH_VEL = 5.000000
HOME_FINAL_VEL = 0.000000
HOME_USE_INDEX = NO
HOME_SEQUENCE = 2

So it was a mistake, that is a huge relief.
And i was not aware who did it as i did not bother to check from frustration.
Thank you.

I can absolutely understand if it was a mistake, but from a bit of reading it does not seem to.

So you think that I broke AxisGUI on purpose to fix something in Gmoccapy? Really?
I'm sorry to say but comments like these are what I would call 'plain absurdity'.
And just to warn you there is going to be yet another release because I screwed up fixing yet another bug.

Ethernet latency is the most common cause of the "Error finishing read"  fault

It basically means you have lost connection to the card, either because too
many read requests went unanswered within in the time limit  (with default settings
that time limit is 80% of the servo thread period and 5 successive requests must time out)

This may be caused by poor Ethernet latency or something more basic like power problems
on the FPGA card, host Ethernet power management issues, or Ethernet hardware problems.


What LEDs are illuminated on the 7I96S when you get the error?

Can you run this command after LinuxCNC gets the error and post the results here:

halcmd show all *.tmax

Also the CPU speed would help.

 

I have disabled all BIOS performance options:

- C-States
- Intel Turbo Boost Technology
- EIST

And so on.

But the error keeps appearing.

I have tried everything. I am quite frustrated with Mesa and LinuxCNC. I knew it wouldn't be easy, but not this complicated!

There's also 7I96S firmware to allow a parallel connected MPG to connected
the  7I96S's P1 expansion connector:

forum.linuxcnc.org/27-driver-boards/4691...ncoder-on-p1?start=0