Advanced Search

there is only to be one var file at any time, it can be named whatever you like, but the ini must call that file name. not knowing what version of probe basic you are on or what method of installation you used makes answering your question very difficult. for that purpose, i would say to remove probe basic and reinstall using the apt installation method. going forward this will update probe basic whenever you run sudo apt update sudo apt upgrade to update your system. any changes of note will appear on the status splash page so you should read the notes thoroughly to find if there are any changes required to update your config files and how to do so. this will put you on an easier path going forward. if you have a working config currently, you should be able to simply review the installation docs for the necessary files your config needs from the probe basic simulation as well as any possible edits to your ini file or existing hal files and subroutines folders. it is advised to always copy over the latest subroutine folder or at the least do a file change comparison with something like meld software to determine what changes were made. unless you are modifying the subroutines the easiest way is to simply copy and overwrite the existing subroutines folder.

my guess is that your installation is likely ahead of your outdated config folder, but again, i can't know what you don't disclose so the only sure way for me to instruct you in resolving is to reinstall probe basic. this also assumes you are on the latest Debian bookworm release and the lockdown version of linuxcnc 2.9.4

if you are unsure, i would recommend uninstalling linuxcnc as well and reinstall using the instructions in the probe basic docs. this will put you on the linuxcnc lockdown deb file install which will only be updated when linuxcnc sends out a lockdown release update. this is what probe basic is tested on and what we support.

hope this helps!

Chris

Hi,

I changed the planner's infrastructure so it can use different smooth algo's easy.
This was quite a lot off work.

Now this is the centroid smooth algo on the left. First iteration.
And a catmull rom spline on the right. Second iteration.
The distance from the corners to the spline is the exact G64 P[x] value.
First and last gcode line are original, not curved.

Does not look like pilot start, there is a missing connection for a usually red thinner wire.
More pictures?

gantry is the old xzero one. 3/4" plate, with 3/4" x 1.5" ribs. basically making a 5"x2.25"x.75" thick "channel". the lower bar connecting the uprights on the bottom is 1x3".

I am adding some reinforcing to prevent racking, you can see it in the last renders.

The "new" minimonster will be all 1" plate boxed with much less gantry span relative to travel, so it will be way stiffer,

  Thanks Rod,

I have used 7/8 - enc, 9 - gnd, 12 - 5v+

Should i change to 6 for 5v+ ??

I see. But surely the files, eg 'change_g43.ngc', have to be somewhere.

Alas, I have no way to check as all my setups are local builds.

assuming the plasma firmware is installed and it is a 7i96, The encoder pins are on TB2 as follows

Well, that looks right.
PCW, the encoder details seem correct as they have been listed in an earlier post
I don't think I can help further..
 

Realised that T17 didn't have the input on the encoder set for arc ok, T18 does.

Double checked wiring, 99% sure it's all correct.

 
 
 

The cx9010 is normally set up an Ethercat master. Linuxcnc uses  a different Ethercat master (and a whole lot more) too.
It would only be useful if you could install Debian and Linuxcnc on it, it could be used but I suspect that will not be possible due to lack of resources

Yes, this can't be right:

setp hm2_7i96.0.encoder.00.scale -1

It also looks almost like the velocity is random numbers
which suggests a wiring error or 7I96 encoder input 'A' not
jumpered for differential mode.

Well the THCAD data is in hal so you are nearly there
what have you got set for 
Voltage scale and voltage offset here?
linuxcnc.org/docs/stable/html/plasma/qtp...lasma:parameters-tab
Maybe PCW can see something missing on the night shift...

i'll have a look at it and try to treat the values as single bits

unfortunately i don't have any test-hardware for it at the moment, but maybe i'll get something made.

btw: where did you have your tangbob boards made ? ready assembled or soldered yourself ? if assembled, what does it cost ready made?

greetings,
Olli

I have connected the boards to the machine, powered it, and got the encoders working and calibrated. Probably ought to throw in a smiley here.
However, moving on, I cannot get the emergency stop system to work as I expected.
When I setup a Output 0 with
net estop-loopback => hm2_7i92.0.7i77.0.0.output-00
Confused myself for a while by not having the field power connected, but now it is I expected that after linuxcnc was launched, and the soft EStop button on the screen was clicked followed by the soft power button, I'd be able to latch the COR relay (In the diagram above and yes the air is on)  by pressing the power button on the machine panel. But it doesn't. 
I guess I am missing something again.

So i have done the calculator, and fudged the numbers to get the voltage back around 0v

Confirm - 1/32 setting on board.
Confirm - Plasma is set to 20:1.
Confirm - torch starts and the light on the THCAD 10 flashes faster.